The effects of serum hepatitis B e antigen (HBeAg) on the prognosis of hepatocellular carcinoma (HCC) patients after hepatectomy remain controversial. Our aim was to explore the prognostic significance of serum HBeAg on the prognosis of patients with HCC using a propensity matching model. Between January 2009 and March 2015, 953 patients with HCC who underwent hepatectomy in West China Hospital were analysed. Propensity matching analysis was applied, and survival analysis was performed using the Kaplan-Meier method. Risk factors were identified by the Cox proportional hazards model. All patients with HCC were classified into an HBeAg(-) group (n = 775, 81.3%) or an HBeAg(+) group (n = 178, 18.7%). Patients with positive serum HBeAg had poorer recurrence-free survival and overall survival before and after propensity matching. Similar results were found in patients within the Milan criteria. For patients beyond the Milan criteria, the HBeAg(+) group had poor overall survival before and after propensity matching. In term of recurrence-free survival, there was no statistically significant impact after propensity matching (P = .055), although there was a trend for HBeAg(+) patient to have reduced recurrence-free survival. Positive serum HBeAg, positive HBV-DNA load, largest tumour size, multiple tumours, microvascular invasion and a high serum level of preoperative alpha-fetoprotein were risk factors for recurrence. Our propensity model confirmed that positive serum HBeAg had a negative impact on the recurrence and long-term survival irrespective of tumour stages. HBeAg seroconversion might be beneficial for reducing the rate of recurrence.