BLAQUE BAUX

Production research interface for the BLAQUE BAUX corpus, Labs, BLOCK, Capstone, and Tracker.

The dealer book capstone (sell-side: sell structure/income/insurance, be the house). CONDITIONAL: steady low-vol carry (+6.9%/yr, Sharpe +0.73) but structurally SHORT CONVEXITY (skew -1.24, bleeds when customers collect); slightly dominated by 60/40 — the real dealer edge isn't replicable in ETFs. Counterparty to bulkhead/barbell. [Capstone]

The liability-driven book capstone (pension/insurer). KEEPER on the funded-ratio mandate: +5.9%/yr surplus over the liability at funded-ratio drawdown -22% vs a naive 60/40's -37%, positive skew, crisis-convex. LDI hedge + spread carry + plain-equity growth + convex tail. Gives up absolute return for the plan. [Capstone]

The glide path (target-date de-risking) — an edge, or just a schedule? CONDITIONAL: no return edge (glide ~= its own average static allocation, timing neutral); value is sequence-of-returns protection — late-horizon drawdown -9% vs -17% static 90/10, shielding the largest balance from a late crash. Horizon risk-management, not alpha. [Conditional]

Annuity / guaranteed-income replication (buffer + bond ladder). CONDITIONAL: the investment side is a cheaply-replicable bond ladder + equity sliver; the ~2.5%/yr VA wrapper more than halves return (DIY +4.6% to +2.0%). The one unreplicable value is the mortality credit. DIY the investment, buy only longevity insurance. [Conditional]

REIT sub-sector rotation — a real edge, or one rate factor? NULL: sub-sectors are one factor (1.2 effective bets, +0.78 corr); a top-2 momentum rotation (+0.4% CAGR) underperforms both equal-weight (+3.9%) and just holding broad REIT (VNQ +5.5%, XLRE +6.8%) net of cost — value-destroying turnover. Confirms brownstone. [Null]

Taleb's barbell — safe core + convex sliver vs the mediocre middle. CONDITIONAL KEEPER: the two-wing barbell (85 cash/7.5 UPRO/7.5 VIXM) dominates a 60/40 on Sharpe (+1.58), Calmar (+0.98), skew (+0.63), maxDD (-4%), at the cost of CAGR. Antifragile signature real (green every crisis). Caveat: cash-heavy flatters Sharpe. [Conditional]

Commodity seasonality — real calendar premium or backtest artifact? Honest out-of-sample test (expanding window, no look-ahead). NULL: in-sample best-months look great but OOS decay to noise/negative; the physical seasonals are real (nat-gas Jan +1.3%/mo, gasoline Apr +5%/mo) but un-harvestable on ETFs — UNG bleeds -22%/yr to contango. [Null]

The hedge fund in an ETF — are HF returns rentable factor beta? (Lo, on QAI/DBMF/MNA/BTAL). CONDITIONAL: multi-strat clone QAI is 75% replicable beta, ~0 alpha, lags a 60/40 — 2-and-20 ate the rest. Honest exception: DBMF managed futures (corr->SPY +0.11, +7.9%), a real diversifier the clone can't capture. Rent DBMF, skip the rest. [Conditional]

The endowment model (Yale/Swensen) replicated in liquid form. NULL: dominated by a boring 60/40 (replica Sharpe +0.57/-35% DD vs 60/40 +0.93/-21%); it's global equity beta (corr->SPY +0.88, 1.4 effective bets) whose 'alts' are levered equity. Can't buy the real edge (illiquidity/manager selection) — why Swensen told retail to index. [Null]