Uncovering the Dark Side of Accounting

Course Director: Ting Dong

Purpose

The purpose of the course is to convey knowledge on how, why, and where accounting goes wrong — and on how such failures are detected in data. The course will pay special attention to areas of financial reporting most dependent on managerial judgment, where material misstatement and outright fraud have repeatedly originated. Another objective is to equip students with a modern, AI-assisted analytical toolset, so that they can work directly with financial statement data and corporate narratives to identify anomalies themselves, without prior programming experience.

Content

Accounting numbers are produced under incentives. Managers face pressure to engage in unethical activities in order to, for example, avoid losses, meet analyst forecasts, and present a credible sustainability profile. Discretion embedded in accounting standards gives them room to respond to that pressure. Understanding this behavior matters both for external stakeholders (investors, creditors, regulators) and from an internal governance and audit perspective. The course combines a fraud analysis perspective with hands-on data analytics, and builds on students' prior knowledge of financial accounting and financial statement analysis.

Issues that will be discussed include, for example:

• Incentives, opportunity and rationalization: the economics of misreporting
• Accounting topics where some of the most complicated accounting scandals are involved, such as related party networks and special purpose entities, revenue recognition and long-term contract accounting, and deferred tax.
• Earnings management and the distributional properties of reported earnings
• Investor and analyst expectation management
• Corporate narratives, ESG disclosure and greenwashing
• Foundations of coding for data analysis: working with financial datasets, cleaning and preparing data
• Reproducible and transparent empirical work with the help of generative AI
• Detection in practice: anomaly and fraud indicators in structured data, textual analysis of unstructured disclosure
• Case analysis of major reporting failures

Format & Structure

The course is built on two interwoven tracks: lectures on accounting fraud and complex reporting areas, and data analysis based on hands-on Python + AI sessions. Each week pairs a 90-minute lecture with a 90-minute seminar. The link between theory, practice, and data is emphasized.

The accounting fraud lectures are followed by case seminars in which students work through an actual reporting failure case. The Data Analytics sessions are workshop-based: students code along, starting from data preparation and progressing to catching earnings management, identifying analyst expectation management, and spotting greenwashing with textual analysis. No prior programming experience is required. We use OpenAI Codex for data analytics sessions.

Assignments run parallel to the teaching and build on each module, and students are expected to prepare for seminars by completing the assignments beforehand. The course concludes with a final seminar, in which student groups present and defend their own analysis.

Course administrator: Christina.Ekelin@hhs.se

On successful completion of the course, students will be able to demonstrate knowledge and practical competence in the following areas:

• Evaluate the incentives, drivers, and mechanisms of financial misreporting, fraud and unethical disclosure
• Interpret and analyze complex accounting areas susceptible to managerial bias
• Use AI-assisted data analytics techniques for analyzing corporate disclosures
• Conduct and present empirical case analyses of corporate reporting prone to manipulation

Students need prior knowledge of financial accounting at an intermediate level. No prior programming experience is required.

The course literature is mainly based on current academic literature on financial disclosure, earnings management, and fraud.

Key readings include:

• Bartov, E., Givoly, D., Hayn, C. (2002). The Rewards to Meeting or Beating Earnings Expectations. Journal of Accounting and Economics, 33 (2): 173–204.
• Burgstahler, D., Dichev, I. (1997). Earnings Management to Avoid Earnings Decreases and Losses. Journal of Accounting and Economics, 24 (1): 99–126.
• Dechow, P. M., Ge, W., Larson, C. R., Sloan, R. G. (2011). Predicting Material Accounting Misstatements. Contemporary Accounting Research, 28 (1): 17–82.
• Degeorge, F., Patel, J., Zeckhauser, R. (1999). Earnings Management to Exceed Thresholds. Journal of Business, 72 (1): 1–33.
• Grewal, J., Richardson, G., Wang, J. (2026). Effects of Mandatory Carbon Reporting on Greenwashing. The Accounting Review, 101 (4): 263–293.
• Larcker, D. F., Zakolyukina, A. A. (2012). Detecting Deceptive Discussions in Conference Calls. Journal of Accounting Research, 50 (2): 495–540.
• Case material and short technical notes for the Python sessions are distributed via the course page.

No textbook purchase is required.

Attendance at the final seminar is mandatory.

Students who get a “pass” in all assignments are eligible to be evaluated for the course grade.

The course grade is based on two parts:

1. Lecture and seminar performance (individual-based, 80% of the total score).
2. Final seminar: group presentation and debate (group-based, 20% of the total score)

Find the course schedule in TimeEdit

Have you been granted educational support because of a documented disability? If so, please contact the course director as soon as possible so that adequate accommodations can be made.

Contact specialneeds@hhs.se if you have questions about being granted educational support or need a copy of your educational support certificate.

For more information, visit the Specific Educational Needs page on the Student Hub (login required).