Corporate Operating Risk – Discussion Notes

---

Current Approach

• Measuring operating risk as standard deviation of YoY revenue growth over 40 quarters
• Using Möbius transformation for symmetry: f(x) = x / (x + 2)
• Including both revenue and cost volatility

---

Möbius Transformation Details

Using parameters (a=1, b=0, c=1, d=2):

f(x) = x / (x + 2)

Properties

• f(0) = 0
• Increasing function
• Maps [-1, ∞) → [-1, 1)
• f(-1) = -1 (respects -100% bound)
• Compresses large positive values

Behavior

• Near zero: approximately linear (slope ≈ 0.5)
• Downside: more spread near -1
• Upside: increasingly compressed

Implication

• Not mathematically symmetric
• Economically meaningful asymmetry (downside bounded, upside unbounded)

---

Key Model Enhancements

1. Detrending

• Remove growth trend before computing volatility
• Use residual-based volatility

2. Downside Risk

• Semi-variance (focus on negative outcomes)
• Compare with transformed volatility

3. Cost–Revenue Interaction

• Compute elasticity: ΔCost / ΔRevenue
• Or correlation between cost and revenue changes

Interpretation:

• High elasticity → flexible cost structure → lower risk
• Low elasticity → rigid costs → higher risk

---

4. Shock vs Structural Volatility

• Use AR(1) model
• Decompose:
  • Persistent variance (cyclical)
  • Residual variance (true uncertainty)

---

5. Time-Scale Effects

• Compare:
  • Quarterly YoY volatility
  • Multi-year (e.g., 3-year CAGR) volatility

---

6. Macro Sensitivity (GDP)

Model: g_firm = α + β g_GDP + ε

Decompose:

• Systematic risk: β × σ_GDP
• Idiosyncratic risk: σ(ε)

---

7. Regime Instability

• Rolling volatility (e.g., 12 quarters)
• Variance of volatility

---

8. Distribution Shape

• Skewness
• Kurtosis (tail risk)

---

Composite Structure

A. Core Uncertainty

• Detrended revenue volatility
• Idiosyncratic volatility

B. Operating Leverage

• Cost elasticity
• Margin volatility

C. Downside / Convexity

• Semi-variance
• Transform vs raw gap

D. Stability

• Volatility of volatility
• Regime shifts

---

Generalization of Möbius Transform

f_k(x) = x / (x + k), k > 1

• Smaller k → stronger curvature
• Larger k → closer to linear
• k = 2 is a balanced choice

---

Practical Recommendation

• Keep current Möbius transform
• Report both raw and transformed volatility
• Add:
  • Cost–revenue elasticity
  • GDP decomposition
  • Volatility-of-volatility

---

Key Insight

The model is already strong; further improvements come from:

• Structure
• Decomposition
• Interpretation

rather than adding more raw metrics.