Skip to main content

Earnings management: detection, application and contagion

Nguyen, Nguyet Thi Minh (2017) Earnings management: detection, application and contagion. Doctor of Philosophy (PhD) thesis, University of Kent,. (KAR id:59823)

Language: English
Download (3MB) Preview
[thumbnail of 64PhD thesis - v51 - minor corrections made - no track change.pdf]
This file may not be suitable for users of assistive technology.
Request an accessible format


The accounting scandals in the 2000s and 2010s have led to a number of large-scale reforms in financial reporting and corporate governance regulations around the world, and still attract a lot of public debates recently. In that context, the demand for further knowledge on earnings management is very topical. What we have known is earnings management does exist. What we have not known, however, seems still overwhelming. We need to know more about issues such as how earnings management could be detected, to what extent earnings management has an impact on investment decisions, what drives earnings management behaviour etc. The accounting research community has responded to such demand by producing a very large, and still growing, volume of publications on the topic during the last few decades. In fact, earnings management has now been one of the largest strands in the mainstream accounting literature.

The first empirical chapter constructs a signal-based composite index, namely ESCORE, which captures the context of earnings management. Specifically, ESCORE aggregates fifteen individual signals related to earnings management based on prior relevant literature. Empirical results using UK data shows that when ESCORE is higher, firms do manage earnings with greater magnitude and are more likely to be most aggressive using both accruals and real earnings management. Firms which are investigated for financial-statement-related irregularities are also shown to have significantly higher ESCORE. The composite score can be easily applied in practice as well as replicated in subsequent studies, especially in emerging market where small samples technically constrain the use of other existing earnings management detection models. The approach to construct ESCORE is innovative and it only measures the likelihood rather than the magnitude of earnings management. This aspect of ESCORE is important given the growing criticisms that none of the existing earnings management models could actually measure the magnitude of earnings management.

The third and last empirical chapter investigates whether aggressive earnings management practices spread across firms sharing interlocked directors. The evidence shows that if a firm aggressively manages earnings (referred to as a 'contagious firm') via accruals (or production activities and discretionary expenses) manipulation in a year, any firms (referred to as 'exposed firms') which are interlocked with that contagious firm in that year and the two following years are more likely to aggressively manage earnings via accruals (or production activities and discretionary expenses, respectively) manipulation. The contagion effect is found to be more pronounced if the interlocked director is male, older, British, and charged with duties which could influence financial reporting. The contagion effect is robust after controlling for endogeneity issues and common characteristics of the interlocked firms. The evidence presented in this chapter is both original and a significant contribution to our knowledge on the impact of board networks on corporate decisions, a topic which attracts a lot of attention as it fits directly to the process of reforming corporate governance codes to enhance the efficiency and effectiveness of the boards of directors.

Item Type: Thesis (Doctor of Philosophy (PhD))
Thesis advisor: Iqbal, Abdullah
Thesis advisor: Shiwakoti, Radha
Uncontrolled keywords: Earnings management, market anomaly, market efficiency, behavioural financem corporate governance, contagion behaviour.
Subjects: H Social Sciences > HF Commerce > HF5351 Business
H Social Sciences > HG Finance
Divisions: Divisions > Kent Business School - Division > Kent Business School (do not use)
Depositing User: Users 1 not found.
Date Deposited: 10 Jan 2017 18:00 UTC
Last Modified: 16 Feb 2021 13:41 UTC
Resource URI: (The current URI for this page, for reference purposes)
  • Depositors only (login required):