A defined-benefit pension is only as safe as its sponsor, yet sponsor distress is under-instrumented relative to plan-side modelling. We present a transparent, reproducible early-warning pipeline built entirely on public data. A Sponsor Vulnerability Index (SVI), a gradient-boosted classifier over point-in-time SEC EDGAR balance-sheet fundamentals and two macro series, predicts a filings-based distress state one to two quarters ahead. On 36 U.S. sponsors in five sector groups (1,418 firm-quarters, 2015Q1-2025Q3) it reaches an out-of-fold AUC-ROC of 0.887 and PR-AUC of 0.711 (base rate 11.9%) under leave-one-firm-out cross-validation. A Monte Carlo VaR with a stylized transition-stress overlay shows that climate-type stress is negligible for a diversified plan but adds 20-43% to one-year 95% VaR when 22.5% of assets are held in sponsor securities. A dynamic self-investment cap maps the SVI to a limit that falls from 8% to below 1% in acute distress and recovers with the sponsor, which we illustrate on PG&E's 2019 bankruptcy and Boeing's persistent fragility. We report the method's failure modes alongside its results (book-equity labels that also fire on non-credit events such as spin-offs, positives concentrated in one firm, and stylized stress parameters) and release code, data and results under a concept DOI.
Figure 1: Left: out-of-fold ROC curve (leave-one-firm-out, AUC =0.887 ). Right: top features by XGBoost gain.
Figure 2: SVI (full-sample refit) for a subset of sponsors, 2015–2025. Distress episodes at PG&E (2018–20), Delta and Occidental (2020–21) and General Electric (2018–19) cross the alert line; Boeing stays above it almost throughout. GE’s second excursion (2023–25) coincides with the spin-offs of GE HealthCare and GE Vernova, a book-equity reduction that is not credit distress. IBM and Walmart stay below the line in this fit.
Sponsor
Positive qtrs
Peak (refit)
Quarter
Peak (out-of-fold)
Quarter
PG&E
8
0.99
2019 Q4
0.99
2020 Q2
Delta
8
0.99
2020 Q4
0.99
2020 Q4
Boeing
42
0.99
2020 Q4
0.96
2020 Q4
General Electric
11
0.97
2018 Q3
0.98
2018 Q3
Coca-Cola
2
0.76
2017 Q4
0.79
2017 Q4
IBM
0
0.62
2015 Q1
0.73
2018 Q4
Table 2: Peak SVI per sponsor: full-sample refit vs. out-of-fold (held-out) scores. Threshold 0.72 .
Sponsor sector, self-investment w
Baseline
Stressed
Add-on
Energy, 22.5%
17.8
25.6
+43%
Energy, 10%
13.5
16.7
+23%
Industrials, 22.5%
15.4
19.1
+23%
Travel, 22.5%
20.0
24.5
+22%
Utilities, 22.5%
12.4
14.8
+20%
Any sector, 0%
11.6–11.7
11.6–11.7
≈0
Table 3: One-year 95% VaR (% of portfolio): baseline vs. stylized transition stress.
Figure 3: Baseline vs. stressed one-year 95% VaR by sponsor sector and self-investment weight.
Sponsor
Peak quarter
Peak SVI
Cap at peak
SVI 2025 Q3
Cap 2025 Q3
PG&E
2019 Q4
0.99
0.9%
0.02
7.9%
Delta
2020 Q4
0.99
0.9%
0.03
7.8%
Boeing
2020 Q4
0.99
0.9%
0.97
1.0%
General Electric
2018 Q3
0.97
1.0%
0.33
5.6%
Exxon Mobil
2020 Q4
0.46
4.7%
0.06
7.5%
Walmart
2020 Q1
0.32
5.7%
0.04
7.7%
Table 4: Dynamic cap (base 10%, mESG=0.8 ) at peak SVI vs. latest quarter (2025 Q3).
Figure 4: Dynamic-cap trajectories (solid) and SVI × 10 (dotted) for PG&E and Boeing, against the static 10% limit and the 8% ceiling implied by mESG=0.8 .