Aims: To examine earning management from agency and signalling theory perspectives. Agency theory was used as a clogging factor for earning management practice whereas, signalling theory relates to managements intention to reflect insider information for the market. Study Design: Considering the nature of the problem, explanatory research design with mixed research approach was employed. Place and Duration of Study: Sample: large manufacturing companies from the period of 2009 to 2017, Addis Ababa, Ethiopia. Methodology: The study used audited financial reports of 14 large manufacturing companies in Addis Ababa operating from the period of 2009 to 2017 for which random effect regression model was used. Results: From agency theory proxies, leverage and audit quality had significant positive and negative impact respectively on earning management. The finding for signalling theory proxies showed that, size of the firm had a positive significant relationship with earning management. Conclusion: The study concluded that signalling and agency theories partially explained earning management in Ethiopian Large Manufacturing Share companies. Originality/value: There were numerous studies explaining earning management from signalling and Agency theory self-reliantly, but this study has modeled earning maneuver motives of management (signalling motive) and controlling mechanisms (Agency theory proxies) set by stakeholders, in one model. Further, the study was conducted in developing country perspective with lower legal requirement on information asymmetry, higher reporting laxity and non-standard/mixed reporting experience.

Aims:To evaluate the strategic role of social and environmental reporting in Ethiopian large tax payer companies based on the resource-based view of the firm. Study Design: The study employed explanatory research design. Place and Duration of Study: Large tax payer companies for the year 2018, Ethiopia. Methodology: The study used annual audited financial reports of 262 companies and structured questionnaire which were distributed to three individuals per each sampled company. Interaction effect model was developed for which ordinary least square (OLS) regression with robust standard errors on a cross-sectional analysis were used to test the hypotheses.