Strategic Control of Facial Expressions by the Fed Chair

View on arXiv ← Back to list

Authors: Hunter Ng

Published: 2024-10-26 16:16:11+00:00

AI Summary

This paper investigates whether Federal Reserve Chairs strategically control their facial expressions during FOMC press conferences and how these nonverbal cues affect financial markets. Using facial recognition technology and deepfakes, the study finds that facial expressions are a distinct signal from verbal content, influencing market reactions differently depending on the chair and their tenure, suggesting investors utilize a dual-processing Markov memory.

Abstract

This article investigates whether the Federal Reserve Chair strategically controls facial expressions during FOMC press conferences and how these nonverbal cues affect financial markets. I use facial recognition technology on videos of press conferences from April 2011 to December 2020 to quantify changes in the Chair's nonverbal signals. Results show that facial expressions serve as a separate public signal, distinct from verbal content. Using deepfakes, I find that the same facial expressions expressed by different Fed Chairs are interpreted differentially. As their tenure increases, negative expressions become more frequent, eliciting adverse market reactions. Furthermore, the markets interpretation of these expressions evolves over time, suggesting that investors process facial cues with dual-processing finite-state Markov memory. In line with the Fed's goals of transparency and non-volatility, I find that Fed Chairs do not strategically control their expressions.

Key findings

The study found that facial expressions serve as a separate public signal distinct from verbal content. The same expressions are interpreted differently across Fed Chairs, and negative expressions become more frequent with tenure, eliciting adverse market reactions, but with decreasing intensity over time. This suggests investors use a dual-processing finite-state Markov memory model when interpreting facial cues.

Approach

The study uses facial recognition technology (DeepFace) to quantify changes in the Fed Chair's facial expressions during FOMC press conferences. Deepfake technology was employed to analyze how the same expressions are interpreted differently across chairs. Regression models were then used to analyze the relationship between facial expressions and market reactions.

Datasets

Videos of FOMC press conferences from April 2011 to December 2020, minute-level market data (SPY, VIX, EUR, JPY), and Google Search Index data.

Model(s)

DeepFace (facial expression recognition), FinBERT (financial sentiment analysis), spaCy (NLP), VGG16 (image recognition), DeepFaceLabs (deepfake generation), and a custom-trained deepfake model.

Author countries

USA

← Previous