1# Contract — F1-conf-032**Question:** association between firm leverage at end of year t and default in t+1, using only info available at end of t.3**Estimand:** b1 in LPM default_next ~ 1 + leverage + cash_ratio + log_sales, OLS on firm-years with observed outcome, SE clustered by firm_id. Meaningful as a conditional association (not causal) assuming the rows are correct firm-year records and linear approximation.4**Outputs:** out/results.json (task_id, n_analysed, n_firms, predictors_used, variables_excluded, beta_leverage, se_leverage_cluster, r2, default_rate, feasibility, feasibility_reason, data_issues, n_input_rows, random_seed); out/analytic.csv; out/exclusions.csv; out/report.md (6 sections); out/analysis.py; out/claims.json.5**Information rules:** predictors only if known at end of t (codebook); rating_2025 known 2025 and built from realised defaults -> must be excluded. Rows with missing default_next excluded and ledgered, no imputation. Only data/firm_panel.csv.6**Invalidity:** using rating_2025; imputing outcomes; counting duplicated rows twice; no observed outcomes.7