Manuscript typeResearch paperResearch aimsWe investigated how corporate governance (CG) affects firm performance (FP) and extends the agency theory by investigating how information asymmetry (IA), proxied by analyst coverage, affects FP in an emerging market.Design/methodology/approach3SLS (Three-Stage Least Squares) regressions were performed on a system of equations based on a panel dataset of 1,768 firm-years of Thai-listed companies from 2014 to 2020 to assess the simultaneous relationships among CG, IA, and FP.Research findingsInternal CG mechanisms (i.e., board size, institutional ownership, and dividend policy) are positively and significantly related to firm performance measured by return on equity (ROE), return on assets (ROA), and price to book ratio/Tobin’s Q (PBR). CG mechanisms are found to reduce IA. Additional analyst coverage was found to increase a firm’s financial performance, ROE by 0.88%, its operational performance, ROA by 0.385%, and its market performance, Tobin’s Q by 0.173 times. Board independence by itself is negatively significant, but interestingly, its interaction effect with analyst coverage is found to be positive and significant on firm performance when measured by ROA.Theoretical contribution/originalityThis study provides further evidence that CG, IA, and FP are simultaneously related and that the interactions between analyst coverage and board independence augment CG and firm performance, particularly via ROA. The impact of agency is significant on firm performance. This study further identifies a ranking of effective CG mechanisms.Practitioner/policy implicationsImprovements in CG channels via board size, institutional ownership, and dividend policy would be associated with greater analyst coverage, which in turn, would provide a monitoring role of the management and reduce principal-agent as well as principal–principal conflicts in addition to adverse selections by stakeholders.Research limitationA seven-year study period may be short to capture a complete business cycle. Survivorship bias is also another limitation of the study
