modeling-clo-cash-flow-waterfalls

name	modeling-clo-cash-flow-waterfalls
description	Builds CLO waterfall models with coverage tests, reinvestment criteria, and distribution allocation across tranches. Use when modeling CLO structures, analyzing OC/IC tests, or projecting tranche returns.
tags	["modeling","structured-finance","investment"]
metadata	{"author":"casemark","practice_areas":["Structured Finance","Securitization","ABS/MBS/CLO"],"document_types":["Financial Model"],"skill_modes":["Modeling","Forecasting"]}

Modeling CLO Cash Flow Waterfalls

When To Use

• Building or auditing a CLO waterfall model from an indenture or offering circular
• Projecting tranche-level returns under base, stress, and recovery scenarios
• Analyzing overcollateralization (OC) and interest coverage (IC) test triggers and cure mechanics
• Evaluating reinvestment period behavior including par-building trades and discretionary trading
• Comparing equity distributions across deal vintages or managers

Inputs To Gather

• Capital structure: Tranche names, ratings, par amounts, spreads/coupons, payment frequency, and stated maturity vs. legal final
• Collateral pool: Par balance, WAS (weighted average spread), WAC (weighted average coupon), WARF (weighted average rating factor), WAL, diversity score, and CCC bucket concentration
• Coverage test triggers: OC and IC trigger levels and cure thresholds for each tranche (typically Class A/B OC, Class A/B IC, etc.)
• Reinvestment criteria: Reinvestment period end date, reinvestment target par, concentration limits, collateral quality tests (minimum WAS, maximum WARF, minimum diversity, CCC cap)
• Default and recovery assumptions: CDR (constant default rate), CPR (constant prepayment rate), recovery rate, and recovery lag in months
• Fee structure: Senior management fee (bps), subordinate management fee (bps), trustee/admin fees, and incentive management fee waterfall position
• Payment dates and day count: Quarterly vs. semi-annual, Act/360 vs. 30/360 [VERIFY against indenture]

Workflow

1. Map the priority of payments from the indenture

   • Build the interest waterfall: trustee/admin fees → senior management fee → Class A interest → Class A OC/IC test → Class B interest → Class B OC/IC test → subordinate management fee → incentive fee → equity distributions
   • Build the principal waterfall: scheduled principal → turbo amortization from failed coverage tests → reinvestment bucket (during reinvestment period) → sequential principal paydown (post-reinvestment)
   • Identify any payment-in-kind (PIK) toggle provisions on mezzanine tranches

2. Build the collateral cash flow engine

   • Project monthly or quarterly interest collections from the pool using WAS, applying CPR and CDR vectors
   • Model defaults: apply CDR to performing par, subtract defaulted par, apply recovery rate after the specified lag period
   • Track performing par balance period-over-period: beginning par − scheduled amortization − prepayments − defaults + recoveries + reinvestment purchases
   • During the reinvestment period, reinvest principal proceeds at the reinvestment target spread subject to collateral quality test constraints

3. Implement coverage tests

   • OC test: (performing par + principal cash) / tranche par outstanding — compare to trigger level each payment date
   • IC test: interest collections (net of senior expenses) / tranche interest due — compare to trigger level each payment date
   • On test failure: divert interest proceeds to accelerate principal paydown of the senior-most tranche until test is cured or tranche is retired
   • Track cure status period-over-period — a cured test restores normal waterfall flow on the next payment date

4. Run the distribution waterfall each period

   • Allocate interest collections top-down per the interest priority of payments
   • Allocate principal collections per the principal waterfall (reinvestment vs. sequential paydown)
   • Calculate equity distributions as the residual after all obligations are met
   • Compute tranche-level IRR, WAL, yield-to-worst, and dollar price for each rated tranche

5. Stress testing and scenario analysis

   • Base case: consensus CDR (e.g., 2%), CPR (e.g., 20%), recovery (e.g., 70%, 12-month lag)
   • Stress case: elevated CDR (e.g., 4–6%), lower recoveries (e.g., 40–50%), extended lag
   • Front-loaded vs. back-loaded default timing vectors
   • Reinvestment spread compression scenarios (e.g., reinvesting at WAS − 50 bps)
   • Sensitivity tables: equity IRR and rated tranche WAL across CDR/recovery matrices

Output

• Waterfall schedule: Period-by-period table showing interest collections, principal collections, fee payments, tranche interest payments, OC/IC test results (pass/fail with ratios), principal paydowns, and equity distributions
• Coverage test dashboard: OC and IC ratios each period with trigger levels, cushion (bps above trigger), and cure/fail status
• Tranche return summary: IRR, yield, WAL, and expected principal window for each tranche across base and stress scenarios
• Equity return profile: Equity IRR, cash-on-cash multiple, and distribution timeline under each scenario
• Sensitivity matrix: Two-variable tables (CDR vs. recovery, CDR vs. reinvestment spread) showing equity IRR and mezzanine tranche breakeven

Quality Checks

• Par balance roll-forward reconciles each period: beginning par − amort − prepays − defaults + recoveries + reinvestments = ending par
• Total cash distributed (fees + interest + principal + equity) equals total cash collected each period — no leakage
• Coverage test ratios match hand-calculated spot checks for at least three periods
• OC test failure correctly triggers interest diversion and principal turbo in the waterfall
• Tranche stated maturities and legal final dates are correctly enforced — no payments after legal final [VERIFY]
• Reinvestment ceases on the reinvestment period end date and the model switches to sequential amortization
• Equity IRR is consistent with market context (broadly syndicated CLO equity typically targets 12–18% unlevered) — flag outliers for review
• CDR, CPR, and recovery assumptions are explicitly stated and sourced; if assumed, mark with [VERIFY]