Effects of Ceo and Director Compensation on Firm Performance

Welcome to our Project!

Abstract:

This study examines the impact of CEO and Director compensation on firm performance. We conducted a regression analysis on a set of determinants derived from academic literature regarding C-suite and Director compensation to determine over or underpayment of these individuals. Our four cases of analysis and hypotheses for them are:

  1. Both CEO and directors are overcompensated: Firm performance will suffer due to backscratching and underperforming executives.
  2. The CEO is overcompensated and directors are undercompensated: Firm performance will excel due to a more qualified CEO and strong controls over the board of directors
  3. The CEO is undercompensated and directors are overcompensated: Firm performance will suffer because the directors are self serving and less motivated to replace the CEO in fear of loosing their compensation
  4. Both CEO and directors are undercompensated: Firm performance will excel because the firm has strong governance to prevent overcompensation

These cases were correlated with a firm performance variable that we created using determinants gathered from a literature search. For analysis, we split our dataset into three bins according to market size. Ultimately, our study found that large firms are negatively impacted by CEO and director overcompensation, medium and smaller firms are positively impacted by it, and smaller firms also have a negative average firm performance associated with underpayment of just the CEO and of both CEO and directors.

Regarding potential sources of error, the datasets gathered from WRDS had missing fields which required either dropping of NaN values in the set or imputing missing variables in our regressions. Further analysis should be done with a larger data set that spans a larger range of years. Note: In this study we used companies listed on the S&P 500.

What is this project?

This is our final project for our Data Science for Finance class at Lehigh Univeristy. Our team, There Is Such A Thing As A Free Lunch, was tasked with creating a research project that utilized all of the skills we learned during the year. In the course, we use python to cover:

  • Pandas
  • Data Cleaning
  • Analysis on Stock Returns
  • Text Scraping and Sentiment Analysis
  • Machine Learning Using Regresion