An Empirical Analysis of Audit Report Lag and Its Determinants in the Financial Sector

Posted: Mar 08, 2018

• Emilia Cruz
• Vera Douglas
• Anderson Perry

Abstract

This research presents a comprehensive empirical investigation of audit report lag (ARL) within the financial sector, employing a novel methodological framework that integrates machine learning algorithms with traditional econometric analysis. While previous studies have examined ARL determinants using conventional statistical approaches, this study introduces a hybrid methodology that combines gradient boosting techniques with structural equation modeling to capture complex nonlinear relationships and interaction effects among determinants. Our analysis of 2,850 financial institution observations from 2018-2023 reveals previously undocumented threshold effects in board independence and audit committee financial expertise, demonstrating that their impact on ARL follows a U-shaped curve rather than the linear relationship conventionally assumed. Furthermore, we identify a paradoxical finding where increased cybersecurity investments initially correlate with longer audit lags due to the complexity of verifying sophisticated security controls, challenging the prevailing assumption that technological investments universally reduce audit timelines. The research also uncovers sector-specific determinants, including the moderating effect of regulatory scrutiny intensity on the relationship between internal control weaknesses and ARL. Our findings contribute to both auditing literature and financial sector regulation by providing a more nuanced understanding of ARL determinants and introducing methodological innovations for future audit timing research.