Report · PCT-027 · Sourced card
Shadow-rate on the agency grid
Maps a privately rated loan onto the agency’s own published grid, factor by factor, so an insurer can see whether the letter rating holds up.
Today vs automated
- Read the methodologyAnalyst · manual
- Copy financialsAnalyst · manual, re-keying
- Score each factorHead of research / IR · manual
- Email / meeting hand-offAnalyst → Head of research / IR
- A privately rated loan is added or re-ratedTrigger: starts on its own
- Read the factor gridAutomated · Data store
- Map and score factorsAutomated · LLM
- Show the indicated rangeAI draft, human review · Reporting template
- The actual opinionApprove / override · Head of research / IR
AutomatedAI draft, human reviewHuman onlyTools: Filing feeds, Data store, LLM, Reporting template
Estimated valueestimate
40–700 analyst hours a year
$6k–$175k of time, vs an assumed $10k–$40k tool cost
Net: −$34k to $165k a year net of tool cost; at the low end of the assumptions it does not cover the tool cost.
Assumptions
- 4–8 analyst hours per borrower today; 1–2 hours with an agent, including review
- 20–100 borrowers a year for one team
- $150–250 per hour loaded cost
- Tool or build cost $10,000–40,000 a year (in-house)
Estimate, based on the assumptions shown; your numbers will differ. How we estimate
What else it can do for the team
- Re-score after each quarter's financials
- Map internal ratings to agency scales
- Prepare for a private letter rating request
- Flag factors closest to a notch change
- Document the mapping for audit
How you get thereFour steps from prompting to an agentic workflow; most teams start at step 2.
Needs: an approved AI tool, a confidentiality policy.
Risk: unchecked answers, data pasted outside policy.
Needs: field definitions, a template, document parsing.
Risk: breaks silently when formats change.
Typical starting point
Needs: inbox/doc-store access, system write-back, logging.
Risk: acting on a misread; needs review gates.
The automated lane above
Needs: shared data model, permissions, audit trail, exception queue.
Risk: errors compound across steps; needs monitoring.
Solution optionsBuy (12 vendors listed), build or hybrid, with the trade-offs of each.
Orion (Private Credit AI), Credit Benchmark, Fitch Ratings, KBRA, Moody's, Lumonic, Hebbia, Lev, Allvue, Chronograph, Cobalt, 73 Strings
Fastest start; you adapt to the product's data model and pricing.
An LLM API, a document parser, a vector or document store, your data warehouse, and a review queue.
Fits your definitions exactly; you own maintenance and controls.
Vendor extraction feeding your own checks, templates and systems.
Common middle path; two things to integrate and govern.
Listing is not endorsement.
What changes for the team4–8 h → 1–2 h per borrower (estimate); coverage goes from the names someone had time to pull to the full set, refreshed when sources change.
How it works
Full title: Shadow-rate a borrower on the agency's own grid, then re-score your internal ratings when financials land
“We scored a public borrower on the agency's own published grid. Here's where it lands.”
This card now includes former PCT-028 (Internal risk rating, re-scored the day the financials land, with the memo written).
Scenario
Public: an insurer holds a privately rated loan. Its credit team wants an independent check of where the borrower's numbers fall on the agency's published scorecard.
What the human still decides
The actual opinion. The output is labeled 'illustrative, not a rating' and is never presented as a rating.
Why it matters
Private letter ratings face scrutiny: NAIC rationale-report rules, SEC questions to Egan-Jones, and a WSJ count of about $40bn of insurer debt Egan-Jones privately rated.
What the record shows
Checked 2026-10-15.
Agencies are publishing more of their private-credit view: Moody's EDF-X Private Credit Model scores one-year probabilities of soft credit events (waivers, amendments, PIK), calibrated on Allvue-contributed borrower data, and Moody's own data show soft events lift the default rate from 4.0% to 5.8%. DBRS has 18% of ~510 middle-market ratings in CCC (high) to C. An internal grade that ignores soft events will look generous next to both.
Try it: prompt and data
Using the attached published methodology, list each scorecard factor and its thresholds. Map this company's 10-K figures to each factor with cites. Output an indicated range labeled 'illustrative, not a rating'.
Data
Public inputs only (SEC/EDGAR filings, published methodologies, public signals). Every figure is cited.
Who uses it, vendors and build
- Segments
- Research & ratings
- Who uses it
- Rating analyst, Insurer / LP credit research, Bank credit
- Build
- Buildable in-house
- In-house
- Methodology parsed once into a scoring template + LLM mapping with cited inputs. Estimate: 1–2 weeks per methodology.
- Last verified
- 2026-10-15
Vendors that claim this capability
- Credit Benchmark: consensus-credit-data
- Lumonic: risk-rating-rescore
- Moody's: credit-ratings
- Orion (Private Credit AI): shadow-rating
- Fitch Ratings: rating-criteria
- 7 more vendors claim related capabilities. See all vendors for this stage →
Tags reflect each vendor's own public claims at the linked page (checked 2026-10-15). Not tested, ranked or endorsed.
Sources and related
- Morningstar DBRS private credit roundtable recap
- Moody's: developing a shared view on private credit risk (EDF-X)
- Private Equity Wire: PIK amendments mask US private credit stress (Moody's)
- Morningstar DBRS: distressed categories now 18% of its private credit ratings; downgrades outpace upgrades 3.1 to 1
- Moody's details EDF-X Private Credit Model, powered by Allvue data, scoring 'soft' credit events
- S&P Global makes RatingsXpress credit ratings available to MCP-enabled AI tools through its AI Data Portal
Next step
Get new walk-throughs
Not investment advice. Names marked SAMPLE are fictional. Found an error? See the corrections log.