Financial conditions and the upper-tail risk of unemployment in the United States

This letter examines the association between financial conditions and the conditional distribution of cumulative changes in the U.S. unemployment rate over each forecast horizon. Using monthly data from 1971 to 2026, we estimate direct quantile local projections over horizons of 1 to 60 months. At 12 months, a one-unit increase in the Chicago Fed National Financial Conditions Index (NFCI) is associated with changes of 0.34, 0.57, and 0.89 percentage points at the 10th, 50th, and 90th conditional percentiles. Median and upper-tail associations are already positive at one and two months. For the median, 75th, and 90th percentiles, the largest point estimates occur at 25–36 months, but their exact locations are sample-sensitive. The upper-tail ordering does not persist at longer horizons. Upper-tail-minus-median intervals remain positive at 12 and 18 months across the six tested block lengths from 6 to 60 months. The estimates describe conditional predictive associations, rather than structural causal effects, and can complement mean-based unemployment monitoring.