WARF (Weighted Average Rating Factor)
Definition
A numerical measure of credit quality in CLO and structured credit portfolios that converts letter ratings (B, BB, BBB) into risk factors, then calculates a weighted average based on each loan's size. WARF is primarily used in Moody's CLO methodology, with lower numbers indicating higher credit quality. For example, a Baa2/BBB rating equals 610, while B2 equals 3490.
Why it matters
WARF is one of the most critical covenant tests in CLO indentures. Managers must keep portfolio WARF below maximum thresholds (typically 2800-3000 for broadly syndicated loan CLOs). Exceeding WARF limits restricts trading and can trigger test failures even without defaults. WARF creates tension between yield maximization (lower-rated loans pay higher spreads) and covenant compliance (lower-rated loans increase WARF). Understanding WARF is essential for analyzing how managers navigate credit cycles—whether they trade up in quality (lowering WARF) or down (increasing yield).
Common misconceptions
- •WARF is not a direct measure of default probability. It's a relative ranking system that maps to historical default rates, but the relationship isn't linear. A WARF of 3000 doesn't mean 30% default probability.
- •Different rating agencies use different systems. WARF is Moody's methodology. S&P uses a different rating factor system with different numerical scales. A portfolio can pass Moody's WARF test but fail S&P's equivalent.
- •Split ratings create complexity. If a loan is rated B1 by Moody's (2720) and B+ by S&P (different scale), the CLO may use the lower rating or an average, depending on indenture language. Verify which methodology applies.
- •WARF can deteriorate without downgrades. If managers rotate from BB loans to B loans (both performing), WARF increases even though no credit events occurred. Market participants sometimes confuse WARF increases with credit deterioration, when it may just reflect strategy changes.
Technical details
Moody's rating factor scale
Moody's assigns numerical factors to each rating: Aaa: 1, Aa1: 10, Aa2: 20, Aa3: 40, A1: 70, A2: 120, A3: 180, Baa1: 260, Baa2: 360, Baa3: 610, Ba1: 940, Ba2: 1350, Ba3: 1766, B1: 2220, B2: 2720, B3: 3490, Caa1: 4770, Caa2: 6500, Caa3: 8070, Ca: 9998, C: 10000. The scale is non-linear—moving from B2 to B3 adds 770 points, while moving from Baa2 to Baa3 adds only 250 points. This reflects accelerating default risk at lower ratings.
WARF calculation methodology
WARF = Σ (Par Value of Each Loan × Rating Factor) ÷ Total Par Value. Example: $500M portfolio with $200M B2 loans (factor 2720), $200M B1 loans (factor 2220), $100M Ba3 loans (factor 1766). Calculation: [(200M × 2720) + (200M × 2220) + (100M × 1766)] ÷ 500M = [544,000M + 444,000M + 176,600M] ÷ 500M = 1,164,600M ÷ 500M = 2,329. Portfolio WARF is 2,329. This is well below typical 2800-3000 limit for BSL CLOs.
WARF limits by CLO type and vintage
Broadly Syndicated Loan (BSL) CLOs: typically 2800-3000 maximum WARF. Middle Market CLOs: typically 3200-3600 maximum WARF (riskier underlying credits). European CLOs: typically 2900-3100 (similar to US BSL). Vintage matters: 2005-2007 CLOs had looser WARF limits (3200+), post-crisis deals tightened significantly (2750-2850 became common). Current market deals (2020-2024) typically 2800-2900 for BSL. Exact limits are indenture-specific and negotiated based on manager track record, target returns, and market conditions.
Impact on manager trading decisions
WARF limits constrain portfolio construction and create strategic trade-offs. Example decision point: Manager sees value in deteriorating BB loan trading at 85 cents (potential recovery to par = 18% return). But adding BB exposure increases WARF. If portfolio is at 2750 (near 2800 limit), buying the BB loan may push WARF over threshold. Manager must either: (1) Pass on the opportunity, (2) Sell higher-rated assets to make room (lowering overall yield), or (3) Accept WARF breach and loss of trading flexibility. This dynamic explains why CLO managers often become forced sellers of deteriorating credits before downgrades—not because fundamentals are terrible, but because WARF constraints force action.
WARF vs actual portfolio credit quality
WARF measures rating distribution, not fundamental credit quality or recovery values. Two portfolios with identical 2500 WARF can have vastly different risk profiles. Portfolio A: 100% B1 loans (WARF 2220), but all in stressed industries with 50% recovery estimates. Portfolio B: Mix of Ba3 (1766) and B2 (2720) loans averaging 2500 WARF, but in diverse industries with 70% recovery estimates. Portfolio B has higher WARF but potentially better fundamentals. This is why sophisticated CLO analysis looks beyond covenant compliance to industry concentration, covenant quality, and manager alpha.
WARF behavior during credit cycles
WARF tends to drift higher during benign credit environments as managers reach for yield (downgrade from BB to B increases spread by 200-300bps). During credit stress, managers typically trade down WARF as they rotate to quality (upgrade from B to BB). March 2020 example: Many CLOs saw WARF decrease as managers sold deteriorating credits and bought higher-rated names, even though market prices suggested credit was deteriorating. This is counterintuitive but reflects WARF mechanics—ratings are sticky and lag fundamentals, so WARF can improve while credit quality actually worsens. Always cross-reference WARF with CCC bucket trends and default rates.