TX (Ternium S.A.) - Quantitative Analysis Executive Summary

Date: December 3, 2025 Framework: Elements of Quantitative Investing

VERDICT: GOOD HYPOTHESIS ✓

Confidence Score: 80% (4/5 quantitative factors passed)

The investment thesis is quantitatively sound for enterprising value investors with 3+ year horizon and tolerance for high volatility.

The Numbers That Matter

Factor Model Decomposition

E(R) = Rf + β·MRP + Value Premium + Quality Penalty + Momentum + Dividend

Expected Return Breakdown:
  Risk-free rate:          +4.50%
  Market premium (β=1.29): +7.74%
  Value factor (P/B 0.61): +6.00%  ← Largest contributor
  Quality penalty:         -2.50%  ← Earnings volatility
  Momentum:                +3.00%
  Dividend (risk-adj):     +4.39%  ← NOT 7.17% due to cut risk
  ─────────────────────────────────
  TOTAL:                   23.13%

Key Insight: The thesis rests on the value factor (+6%) and cyclical recovery assumptions. If earnings don't recover, the expected return drops to ~12% (still acceptable but not exceptional).

Three Critical Disagreements with Original Report

1. Dividend Income is Overstated ❌

Report Assumes: 7.17% dividend yield maintained Quantitative Reality: 4.39% expected yield

Why: Payout ratio of 106.8% is mathematically unsustainable. Scenario analysis:

• 20% probability: No cut
• 35% probability: 25% cut
• 30% probability: 50% cut
• 15% probability: Full elimination

Impact: Expected return is 2.78% lower than report implies from dividend alone.

2. Tail Risk is Understated ❌

Report Claims: "20-25% maximum drawdown" Quantitative Reality: 50-70% drawdowns possible (9-15% probability)

Why: Steel sector volatility is 35% annually, not 20%. Monte Carlo simulation shows:

• 9% probability of >50% loss over 3 years
• CVaR (95%): -49% loss in worst 5% of outcomes
• Cantelli's inequality: P(Loss > 70%) ≤ 13.5%

Impact: Psychological preparation required for brutal drawdowns.

3. "Base Case" is Actually Optimistic Scenario ❌

Report: "Reaching Graham Number ($63.16) is base case" Quantitative Reality: 31.6% probability = 70th percentile outcome

True Base Case (Median): +27% over 3 years (~8.5% annualized)

Why: Monte Carlo shows fat left tail from cyclical extension scenarios. The Graham Number requires:

1. EPS recovery from $2.90 to $7.34 (153% growth)
2. P/B ratio expansion from 0.61x to 1.0x
3. No dividend cut (20% probability)

All three must occur—unlikely to be "base case."

Three Critical Agreements with Original Report

1. 5% Position Size is Correct ✓

Kelly Criterion suggests: 87% allocation Report recommends: 5% Quantitative Verdict: 5% is prudent

Why: Despite high expected return, the following risks justify capping position:

• 35% volatility (high idiosyncratic risk)
• 80% dividend cut probability
• Earnings quality issues (Information Ratio 0.416 < 0.50 benchmark)
• Model uncertainty for cyclical timing

Lesson: Sometimes qualitative judgment should override quantitative optimization. This is one of those cases.

2. The Balance Sheet is Genuinely Strong ✓

Report: "Fortress balance sheet" Quantitative: Debt/Equity 13.9%, Current Ratio 2.46

Why it matters:

• Prevents bankruptcy even with earnings collapse
• Provides downside protection (not price protection, but solvency protection)
• Explains why TX survives -88% earnings misses

Caveat: Strong balance sheet protects against permanent loss (bankruptcy), NOT quotational loss (price volatility). You'll still experience -40-50% drawdowns.

3. Cyclical Recovery Timing is Plausible ✓

Report Claims: Recovery 2026-2027 Quantitative Model: 63-78% probability of mean reversion within 2-3 years

Why:

• Steel cycles average 3-5 years
• Current downturn began 2023 (~2 years ago)
• ROE reversion from 4.9% to 15% (normalized) has 2-year half-life
• Forward P/E of 5.13x prices in recovery expectations

Caveat: 22-37% probability of NO recovery in that timeframe. This is non-trivial downside risk.

Risk-Adjusted Performance

Sharpe Ratio: SUPERIOR to Market

Interpretation: TX offers 0.532 units of return per unit of risk—60% better than the market's 0.333. This validates the investment thesis from a mean-variance optimization perspective.

Information Ratio: BELOW Institutional Quality

• Alpha: 10.89% (stock-specific return above CAPM)
• Tracking Error: 26.19% (idiosyncratic volatility)
• Information Ratio: 0.416 (benchmark: >0.50)

Interpretation: TX generates positive alpha, but with excessive idiosyncratic risk. The IR of 0.416 reflects the earnings volatility problem—you're not being adequately compensated for the stock-specific risk.

Implication: This is a high-risk, high-return bet, not a "high-quality compounder."

Monte Carlo Simulation Results (10,000 trials, 3-year horizon)

Return Distribution

Key Probabilities

Quantitative Recommendations

1. INVESTMENT DECISION: PROCEED (with caution)

Rationale:

• Sharpe 0.532 > Market 0.333 ✓
• Expected return 23.13% is attractive ✓
• 65.5% probability of profit ✓
• Strong balance sheet prevents bankruptcy ✓

But prepare for:

• Dividend cut (80% probability)
• -40-50% drawdowns (possible in bad years)
• 3+ year holding period required

2. POSITION SIZING: 5% MAXIMUM

Do not exceed 5% even if your conviction is high.

Rationale:

• Kelly suggests 87%, but model uncertainty is high
• Idiosyncratic risk (26% volatility) limits concentration
• Dividend cut will test your resolve
• Cyclical timing can be wrong (22-37% probability)

Alternative sizing:

• Aggressive investor: 5.0%
• Neutral investor: 5.0% (as per report)
• Conservative investor: 0% (failed defensive criteria)

3. ENTRY STRATEGY: Two Tranches

Option A (Tactical Patience):

• Wait for pullback to $34-36 (from current $37.68)
• Enter full 5% position
• Margin of safety increases to 45-50%

Option B (Immediate Entry):

• Tranche 1: 2.5% at current $37.68
• Tranche 2: 2.5% at $34-36 if pullback occurs
• Average cost: $35-36

Quantitative Preference: Option A (wait for pullback)

• Improves Sharpe from 0.532 to ~0.600
• Reduces VaR from -34% to -28%

4. STOP LOSS: $32.00 (Non-Negotiable)

Represents:

• -15% from current price $37.68
• Below 200-day MA ($31.97)
• Signals cyclical recovery thesis is failing

If stop is triggered:

• Exit full position immediately
• Do NOT "average down"
• Reassess thesis with fresh eyes

5. EXIT STRATEGY (Upside Targets)

Rationale: Lock in gains progressively as valuation normalizes.

The Quantitative Truth

What the Math Says

Expected Return: 23.13% annualized Risk-Adjusted: Sharpe 0.532 (excellent) Probability of Success: 65.5% over 3 years

Quantitative Verdict: The numbers support a 5% allocation for value investors who can handle volatility.

What the Math Warns

Volatility: 35% annual (brutal drawdowns) Dividend Cut: 80% probability (income thesis broken) Tail Risk: 9% probability of >50% loss Earnings Quality: Poor (IR 0.416)

Quantitative Warning: This is not a "margin of safety" investment—it's a leveraged cyclical value bet.

The Psychology Test

Can you handle:

1. Watching the position drop -40-50% in year 1?
2. Dividend being cut 50-100% (near certain)?
3. Waiting 3+ years for recovery that might not come (22-37% risk)?
4. Sleeping through 35% annual volatility swings?

If YES to all four: Proceed with 5% position If NO to any one: Reduce to 3% or skip

Remember: The math is compelling, but behavioral finance matters. If you'll panic-sell at -40%, the expected 23% return is irrelevant—you'll lock in losses instead.

Final Quantitative Verdict

GOOD HYPOTHESIS (B+ Grade)

Why B+, not A:

• Earnings volatility (Q3 -88% miss)
• Dividend sustainability issues (106.8% payout)
• Quality factor is negative (ROE 4.9%)
• High idiosyncratic risk (IR 0.416)

Why B+, not C:

• Superior Sharpe ratio (0.532 vs 0.333)
• Strong value factor exposure (P/B 0.61)
• Cyclical recovery probability 63-78%
• Balance sheet protects downside

The Synthesis: If you can accept the volatility and dividend cut, the risk-adjusted returns justify a 5% position. But this requires discipline—use that $32 stop loss, don't exceed 5% sizing, and prepare psychologically for drawdowns.

References

Quantitative Framework: Elements of Quantitative Investing

• Chapter 2: Univariate Returns (volatility estimation)
• Chapter 3: Performance Metrics (Sharpe ratio, expected return)
• Chapter 4: Factor Models (multi-factor decomposition)
• Chapter 5: Risk Evaluation (VaR, Cantelli's inequality)
• Chapter 9: Portfolio Management (Kelly Criterion, position sizing)

Key Formulas Applied:

1. Factor Model: r_TX = α + β·f + ε
2. Sharpe Ratio: SR = (E[R] - Rf) / σ = 0.532
3. Kelly Criterion: f* = (E[R] - Rf) / σ² = 174.78% → Half-Kelly = 87.39%
4. VaR: E[R] - z_0.95·σ = -34.44%
5. Cantelli's Inequality: P(Loss > Lσ) ≤ 1/(1+(L+SR)²)

Files Generated

1. Full Analysis: /home/pengacau/pasar-malam/output/tx-quantitative-verdict-2025-12-03.md (18,000 words)
2. This Summary: /home/pengacau/pasar-malam/output/tx-quantitative-summary-2025-12-03.md (2,500 words)
3. Metrics JSON: /home/pengacau/pasar-malam/output/tx_quantitative_metrics.json
4. Analysis Script: /home/pengacau/pasar-malam/scripts/tx_quantitative_analysis.py
5. Original Report: /home/pengacau/pasar-malam/output/tx-2025-12-03.md

Bottom Line: The quantitative analysis validates the TX investment thesis with a 5% position size for value investors who can handle high volatility. Expected return of 23.13% and Sharpe of 0.532 are compelling. However, be prepared for dividend cuts (80% probability) and brutal drawdowns (35% volatility). Use the $32 stop loss religiously.

Quantitative Grade: B+ (Good Hypothesis)

Graham would say: "The margin of safety is there, but only for the enterprising investor."

The quant says: "The Sharpe ratio is 0.532. The math checks out. But don't exceed 5%, and don't ignore that stop loss."

Analysis Date: December 3, 2025 Analyst: Senior Quantitative Investment Analyst Confidence Level: 80% (4/5 factors passed)