Tax Avoidance Predictions: CETR and Social Network Synergy
Abstract
This study aims to address the challenge of detecting tax avoidance by developing a robust predictive model that integrates social network characteristics with financial metrics. Utilizing the Cash Effective Tax Rate (CETR) as a primary indicator, this research investigates the potential enhancement of predictive accuracy through the inclusion of board director relations and accounting firm affiliations. The methodology involves analysing data from publicly listed companies in Indonesia, spanning five years from 2018 to 2022. Key findings indicate that incorporating social network attributes improves the model's ability to predict tax avoidance. The results reveal that companies with directors serving on multiple boards and those sharing accounting firms tend to exhibit similar tax avoidance behaviours. These insights suggest that social networks play a crucial role in corporate tax strategies, providing a deeper understanding of the dynamics influencing tax avoidance. The conclusions underscore the importance of considering both financial and social network data in developing more effective regulatory frameworks to combat tax avoidance.
Description
Masteroppgave(MSc) in Master of Science in Business Analytics, Handelshøyskolen BI, 2024