analyzing-structured-products

name	analyzing-structured-products
description	Evaluates ABS, MBS, CLO, and CDO structures with cash flow waterfall and subordination analysis. Use when analyzing structured products, modeling cash flow waterfalls, or evaluating tranche risk.
tags	["analysis","fixed-income","risk"]
metadata	{"author":"casemark","practice_areas":["Fixed Income","Credit Research","Bond Trading"],"document_types":["Analysis Report"],"skill_modes":["Analysis"]}

Analyzing Structured Products

When To Use

• Evaluating a new ABS, MBS, CLO, or CDO offering for investment suitability
• Assessing tranche risk and credit enhancement adequacy on an existing position
• Modeling cash flow waterfalls under base, stress, and default scenarios
• Comparing subordination levels, excess spread, and overcollateralization across deals
• Reviewing deal documents (indenture, offering circular, trustee reports) for structural features
• Monitoring portfolio holdings for collateral deterioration or trigger breaches

Inputs To Gather

• Deal documents: Offering circular/prospectus supplement, indenture, servicing agreement
• Collateral data: Loan-level tape or pool stratification (WAC, WAM, WALA, FICO distribution, geographic concentration, LTV distribution for MBS; industry/rating distribution for CLO/CDO)
• Tranche structure: Par amounts, coupon types (fixed/floating), spread, legal final maturity, expected maturity, payment priority
• Credit enhancement details: Subordination percentages, overcollateralization (OC) targets and triggers, excess spread, reserve accounts, insurance wraps if applicable
• Performance data: Monthly trustee/remittance reports — delinquency buckets (30/60/90+), CDR, CPR, loss severity, cumulative losses vs. trigger levels
• Rating agency presale reports or surveillance commentary (Moody's, S&P, Fitch, KBRA, DBRS)
• Pricing/spread context: Comparable deal spreads, secondary trading levels, index benchmarks (e.g., CMBX, ABX, Markit CLO indices)

Workflow

1. Classify the structure

   • Identify product type: RMBS (agency/non-agency), CMBS, auto ABS, credit card ABS, student loan ABS, CLO, CDO (cash/synthetic), or bespoke
   • Note static vs. revolving/managed pool; amortizing vs. bullet tranches
   • Identify waterfall type: sequential pay, pro rata, modified pro rata with trigger-based switches

2. Map the cash flow waterfall

   • Trace payment priority from senior through mezzanine to equity/residual
   • Identify interest waterfall vs. principal waterfall (many deals separate these)
   • Document trigger events: OC tests, IC (interest coverage) tests, delinquency triggers, cumulative loss triggers
   • Note what happens on trigger breach (e.g., diversion of excess spread, turbo amortization of seniors, trapping of equity distributions)

3. Assess credit enhancement

   • Calculate current subordination for each tranche: (total par below tranche) / (total deal par)
   • Compare initial vs. current subordination — has it grown (sequential pay building credit enhancement) or eroded (losses eating into junior tranches)?
   • Quantify OC cushion: (collateral par − tranche par) / tranche par for each OC test
   • Evaluate excess spread: gross WAC − senior coupon − servicing fee − expected losses
   • For CLOs: check reinvestment period remaining, WARF, WAL, diversity score, CCC bucket limits [VERIFY: specific threshold levels vary by deal and manager]

4. Run scenario analysis

   • Base case: Forward curves, current prepayment/default/severity assumptions
   • Stress case: Elevate CDR by 2–3x, increase loss severity 10–20 pp, slow prepayments (extend duration) or spike prepayments (reduce excess spread capture)
   • Tail risk: Model what default rate/severity combination breaks each tranche — solve for breakeven loss multiple
   • For CMBS: model individual large-loan default scenarios and balloon extension risk
   • For CLOs: model par erosion from downgrades/defaults reducing the OC cushion

5. Evaluate collateral quality

   • Review pool stratification for concentration risk (single obligor, industry, geography)
   • Assess vintage performance: compare deal's cumulative loss curve to similar-vintage benchmark curves
   • For MBS: analyze LTV distribution, documentation type, occupancy status, loan purpose
   • For CLOs: review collateral manager track record, portfolio turnover, CCC-rated bucket size, weighted average rating factor (WARF) trajectory

Output

• Structure summary: Deal name, collateral type, closing date, pool balance, number of tranches, reinvestment period (if applicable)
• Waterfall diagram: Simplified schematic showing payment priority, trigger levels, and current OC/IC cushions
• Tranche-level analysis table: For each tranche — rating, coupon, current par, subordination %, OC cushion, WAL, spread, breakeven loss multiple
• Scenario results: Cash flow projections under base/stress/tail showing principal losses, WAL extension, and yield impact per tranche
• Collateral health scorecard: Key pool metrics with trend arrows (improving/stable/deteriorating) vs. prior periods
• Risk flags: Trigger proximity, concentration risks, servicer concerns, rating watch items
• Recommendation: Buy/hold/sell with spread target and key monitoring triggers

Quality Checks

• Verify that waterfall mapping matches the indenture — payment priority errors cascade through all downstream analysis
• Confirm subordination math: tranche sizes must sum to total deal par; percentages should cross-check against trustee reports
• Validate scenario assumptions against historical analogs (e.g., 2007–2009 loss curves for stress cases) [VERIFY: appropriate stress benchmarks depend on asset class and vintage]
• Ensure collateral data vintage matches the most recent trustee report date — stale data understates risk
• Cross-check ratings against all agencies covering the deal; note any split ratings
• Flag any structural features that could impair analysis accuracy: clean-up calls, optional redemption provisions, servicer advancing mechanics [VERIFY: advancing conventions differ by asset class and servicer]
• Confirm that breakeven calculations account for timing of losses, not just cumulative magnitude