milliman pfi macro signal
What it checks
Once a month Milliman publishes the funded ratio for the 100 biggest US corporate pensions. When that ratio drops sharply (often because interest rates fell), pension-heavy companies face bigger cash contributions and EPS drag. Short the basket for a few months.
Mechanism
Milliman publishes a monthly aggregate funded ratio for the 100 largest US corporate DB plans. Sharp monthly funded-ratio deteriorations (driven by rate moves making liabilities expand faster than assets) precede contribution acceleration and EPS drag for the legacy-DB basket — autos, airlines, defense, industrials, legacy tech. The 10-K-parsed per-ticker pension_funding signal (#203) catches the firm-level dispersion; this macro layer catches the system-wide shocks the per-firm series under-samples between annual filings.
Signal rule
PFI funded_ratio_pct month-over-month delta own-z over a 36-month rolling window <= -1.0 (and -1.5 variant) AND ticker in pension-heavy basket -> SHORT on T+5 (publish lag); hold 21 or 63 trading days. Outside the basket, generator returns [].
Data dependencies
daily_pricesAdjusted-close OHLCV for every US-listed ticker; primary price feed.
milliman_pension_fundingWorker data table — see services/worker schema.
Expected edge
- Paper alpha
- ~3-5% over 1-3mo
- Paper window
- T+5 to T+63d
~3-5% over 1-3mo on the pension-heavy short basket during PFI deterioration shocks (analog to Franzoni-Marin 2006 per-firm 12mo drift).
Example tickers where this is likely to fire
Illustrative only — the signal fires based on the live data, not a fixed list.
Related families
pension funding status driftQualityUnderfunded pensions (<80%) → 12m underperformance via earnings drag from contributions.
credit spread shockMacroGilchrist-Zakrajsek 2012 AER show the excess-bond-premium component of credit spreads predicts equity returns at 1-3mo horizons with R² up to 12%. We use raw BAA-10Y (Moody's BAA corp minus 10Y Treasury) from FRED as the cheap proxy. Widening shock = risk-off pressure (short high-beta names with downtrend); sharp compression = risk-on regime (long beta names with uptrend).
altman z scoreQualityAltman's Z-score combines five accounting ratios (working-capital/TA, retained-earnings/TA, EBIT/TA, market-cap-equity/total-liab, sales/TA) into a discriminant score that separates safe (Z>2.99) from distressed (Z<1.81) firms. The market-cap-equity X4 variant (vs Altman's original book-equity) is a stronger forward predictor and is the modern usage. Hilscher-Wilson 2017 J. Banking confirm the score still discriminates default risk in 2000s data.
yield curve sector rotationMacroYield-curve slope (10y - 2y) gates cyclical vs defensive sector longs.
Explore milliman pfi macro signal on alphactor.ai
See which tickers this family is currently firing on, with live signals and rankings.