Structural shifts are undermining traditional leading economic indicators
From late 2021 through 2025 major leading signals signaled recession while jobs, income and spending held up, suggesting pandemic, demographic and market changes have distorted decades-old forecasting tools.
- From late 2021 to 2025 the Conference Board LEI, RecessionAlert USMLEI and a prolonged yield curve inversion repeatedly signaled recession while coincident data showed expansion.
- High confidence distortions include mass boomer retirements, AI-driven wealth concentration that concentrates spending, pandemic-era deposit and savings gluts, and an immigration surge that temporarily altered labor supply.
- The yield curve failed as a predictor this cycle because banks held excess deposits, QE and global reserve demand suppressed long yields, and forward guidance embedded expected rate cuts into long-term rates.
- Leading indices are biased toward manufacturing and sentiment metrics and thus misread a services-dominated, K-shaped economy where the top 10 percent now drive roughly half of consumer spending.
- Measurement and data quality issues, including agency staffing losses and increased statistical imputation, have reduced confidence in several official series.
- Practical takeaway: use multiple indicators, separate temporary from persistent distortions, and recalibrate models rather than relying on any single historical signal.