base rate:
The question is essentially 'will 2026 PBS layoffs exceed 2025 PBS layoffs?' In any given year, layoffs can go up or down — the base rate is roughly 50/50. Looking at the JOLTS PBS layoffs series historically, there's no strong secular trend (they fluctuate around a level that shifts with economic cycles). So the starting point is ~50%.
evidence updates:
Key upward pressures: (1) January 2026 layoff announcements at highest since 2009 (+5%), (2) continued major tech layoffs (Amazon 16K, Block 4K+) suggesting AI restructuring wave continues (+3%), (3) tariff shock creating economic uncertainty (+3%), (4) structural shift toward involuntary separations seen in 2025 JOLTS data (+2%). Key downward pressures: (1) February 2026 total JOLTS layoffs unchanged and DOWN 146K YoY (-5%), (2) December 2025 PBS separations decreased significantly (-3%), (3) overall layoff rate at 1.1% is only modestly elevated (-2%), (4) payroll growth still positive suggesting no imminent recession (-2%). Net: roughly neutral, slight downward pressure of ~2% from 50%.
combination method:
Weighted average of three sub-questions (recession probability 35%, PBS trajectory 40%, AI displacement 25%) with upward adjustment for positive correlation between sub-questions.
final:
The evidence is genuinely mixed. The most critical data point — PBS-specific monthly layoffs for early 2026 — is missing. Total economy JOLTS data suggests no acceleration in layoffs, which is a moderately strong signal against YES. But Challenger announcement data and major tech layoffs suggest continued white-collar displacement. Tariff shock adds upside risk. The lack of PBS-specific data is the largest uncertainty. I land at 48% — essentially a coin flip with a very slight lean toward NO, driven primarily by the February JOLTS total economy data showing no acceleration. Confidence is LOW because the most relevant data (PBS-specific JOLTS monthly figures for 2026) is unavailable.
ensemble:
{'ensemble_count': 3, 'perspectives': {'base_rate': {'p_yes': 0.48, 'confidence': 0.3}, 'evidence_driven': {'p_yes': 0.47, 'confidence': 0.3}, 'contrarian': {'p_yes': 0.45, 'confidence': 0.3}}, 'spread': 0.03, 'individual_p_yes': {'base_rate': 0.48, 'evidence_driven': 0.47, 'contrarian': 0.45}}