Causal analysis of macroeconomic shocks on financial markets through machine learning methods

Campita, Stefano
Benedetto, Francesco

Journal of Economics and Business Letters
Vol. 6 No. 2 (2026): April 2026

Macroeconomic announcements often trigger sharp market reactions; however, their causal impact is difficult to measure. This study quantifies the causal effects of the consumer price index (CPI), non-farm payrolls (NFP), and Federal Open Market Committee (FOMC) decisions on the S&P 500, Gold, and the VIX using daily data from 2022 to 2024. Three estimators are applied: Ordinary Least Squares, Propensity Score Matching, and Double Machine Learning. The results show limited price adjustments but strong and statistically meaningful volatility responses. FOMC shocks generate the most persistent effects, whereas CPI and NFP impacts are short-lived. Overall, the findings indicate that volatility, rather than prices, is the primary transmission channel of macroeconomic news, highlighting the value of causal machine learning in identifying structural market responses.