What (One-liner)

Derives 5 quantitative management signals from yfinance statements + company info (ROIC / revenue per employee / growth consistency / …) plus 5 heuristic qualitative assessments (founder CEO / PE-KE / leadership style / compensation / tenure), producing a 0-100 L score.

How to use

Drag blcv_leadership_v1 onto the Canvas and fill symbol. Output is a quantitative dict + qualitative list (5 items) + l_score + logic_lineage.

Typical usage:

• symbol=META → founder_ceo active (Zuckerberg), ROIC + growth signals from yfinance → l_score ~75-85
• symbol=AAPL → founder_ceo inactive (Jobs long since succeeded by Cook), high ROIC → l_score ~60-70

Core formulas

Quantitative: 5 signals derived from yfinance's three statements + fullTimeEmployees.

Qualitative: 5 fixed dimensions:

1. founder_ceo — lookup in a static 8-symbol table
2. pe_ke — 2×2 matrix based on (growth > 15%, ROIC > 12%)
3. leadership_style — growth-variance threshold
4. compensation — placeholder
5. tenure_stability — placeholder

Scoring: six _add items with weights (2.0, 1.0, 1.5, 0.5, 1.0, 0.5); max_pts = 6.5; l_score = clamp(score/max_pts, 0, 1) × 100.

Assumptions & applicability

Assumptions: yfinance financials available + standard Yahoo naming + the static founder-CEO table is a reasonable proxy + heuristic thresholds 15% / 12% are reasonable for US equities.

Fits: US-listed companies; management-quality support for long-hold candidates.

Does not fit: non-US / non-mega-cap / private companies / < 2 years since IPO / management-transition periods.

Input / Output contract

{symbol: str} → {symbol, name, quantitative, qualitative, l_score, logic_lineage}. The 5 qualitative dimensions come in a fixed order — pinned by the golden.

Known limitations

1. Static founder-CEO table (8 US mega-caps, hand-maintained, annual review)
2. compensation / tenure_stability are placeholders ('data_pending')
3. The growth_consistency heuristic is not an academic convention
4. Scoring weights have had no sensitivity analysis
5. earnings_data is flagged available but unused in scoring
6. PE/KE thresholds 15% / 12% hardcoded

References

Graham-Dodd (1934) + Bertrand-Schoar (2003) manager-style evidence + Fahlenbrach (2009) founder-CEO evidence + Buffett's shareholder letters. Full citations in the frontmatter.

Golden Test

tests/golden/fixtures/tier2/blcv_leadership/ — a yfinance-snapshot integration regression, 1e-6 tolerance post-rounding, 6 tests:

• matches_snapshot — deep comparison of the full output (recursive numeric tolerance)
• l_score_in_0_100_range — range contract
• determinism — 5 reruns byte-identical
• top_level_field_set — 6 fixed top-level keys
• qualitative_has_5_fixed_dimensions — 5 dimensions in fixed order
• founder_ceo_lookup — static-table contract (META/NVDA/TSLA present, case-sensitive)

The snapshot is shared with tier1/blcv_financials (the same AAPL 2026-04-20 pickle).

Changelog

• 1.0.0 (2026-04-20) — First Active (Tier 2 batch 3, yfinance snapshot regression)