Exposing Greenhushing's Impact on CEO Pay

Overview

This repository contains the data, code, and analyses from a research project investigating the relationship between greenhushing—where companies underreport their environmental impact—and excess executive compensation. By utilizing AI and Natural Language Processing (NLP) techniques, we compare corporate disclosures against Sustainability Accounting Standards Board (SASB) guidelines and analyze how the intensity of greenhushing affects CEO pay.

Contents

• Board Data: Contains data related to internal governance factors such as Board Size and CEO’s Power Status.
• CDP Data: Dataset from the Carbon Disclosure Project, detailing environmental performance disclosures.
• ESG Data: Environmental, Social, and Governance (ESG) scores and related metrics.
• Economic Determinants Data: Data on the economic determinants of CEO pay.
• SASB Standards: Guidelines from the Sustainability Accounting Standards Board used in the analysis.
• Compensation.xlsx: Spreadsheet detailing excess executive compensation estimation using linear regression framework.
• Excess GH1.xlsx & Excess GH2.xlsx: Excel files containing data on excess CEO pay & greenhushing regression metrics.
• Final Sample.xlsx: The final sample used in our analysis.
• pearson_correlation_final_sample.xlsx: Correlation analysis of the final sample.
• summary_descriptive_stats.xlsx: Summary of descriptive statistics from the dataset.
• Project Code.ipynb: Jupyter Notebook containing the Python code for data processing, analysis, and modeling.
• Research Report.pdf: The full research paper draft detailing our findings, methodology, and conclusions.

Methodology

1. Data Collection: We aggregated data from various sources, including corporate disclosures, SASB standards, and ESG scores.
2. Natural Language Processing (NLP): Applied techniques like Latent Dirichlet Allocation (LDA) and Continuous Bag of Words (CBOW) and more to analyze the text from corporate reports.
3. Analysis: Examined the relationship between greenhushing and excess CEO compensation, considering internal governance factors and economic determinants.

Key Findings

• Greenhushing significantly increases excess CEO compensation.
• Internal governance factors, particularly Board Size and CEO’s Power Status, influence the extent of this relationship.
• The economic determinants of CEO pay add further complexity to the greenhushing dynamic.

How to Use

1. Clone the Repository:

   git clone https://github.com/Yosri-Ben-Halima/Leveraging-NLP-to-Quantify-Greenhushing-and-Studying-its-Impact-on-Excess-CEO-Pay.git

2. Install Required Packages: Ensure you have Python 3.x and install necessary libraries:

   pip install -r requirements.txt

3. Run the Jupyter Notebook: Explore the data and replicate our analysis by running Project Code.ipynb.

Connect with Me

Thank you for visiting my GitHub profile! Feel free to reach out if you have any questions or opportunities to collaborate. Let's connect and explore new possibilities together!