Purpose: The main therapy for small hepatocellular carcinoma (HCC) is surgical resection, but other treatments including percutaneous ethanol injection (PEI) and radiofrequency ablation (RFA) may have a role. We sought to evaluate the cost-effectiveness of surgical resection and local ablation in the initial treatment of solitary small (≤3cm) HCC in HCV-related hepatitis and cirrhosis. Methods: To simulate the process from first treatment to death from liver disease or other causes, we created a state-transition Markov model which represented the principal health states of chronic hepatitis, cirrhosis, and decompensated cirrhosis with or without HCC. We assumed that the treatment choice corresponded to HCC state features such as maximal size, number and location, that treatment would start with resection or local ablation, including PEI and RFA, and that, as HCC progressed, local ablation plus transarterial chemoembolization (TACE), TACE, hepatic arterial infusion chemotherapy (HAIC) and conservative treatment would be applied. We retrospectively reviewed 326 HCC cases admitted to our institution, including 93 of small solitary HCC, and, for each treatment strategy, estimated the rate of complete remission (CR) and the rates among CR cases of HCC-free survival and treatment for recurrent HCC . We also estimated the rates of transition between treatments in non-CR cases. We applied data on health insurance reimbursement to estimate the annual cost of each health state and the admission cost of each treatment. Results: Our model was validated and consistent with the 5-year survival rate from Kaplan-Meier's analysis of a sample population at our institution and of a nationwide survey of Japan. For patients hypothetically identical with those of the sample population regarding mean age (65 years old), gender ratio (73% male) and cirrhosis rate (93%), surgical resection increased lifetime treatment cost by about $14,000 and extended average survival by about 1.4 years, yielding an incremental cost-effectiveness rate of $10,200 per life-year compared with local ablation therapy. The sensitivity analysis showed that only if the CR rate after resection was less than 0.57 or the death rate following resection was more than 0.1 did incremental cost-effectiveness exceed the threshold of cost-effectiveness. ($50,000/LY) Conclusions: Our analysis suggests that surgical resection is more cost-effective than local ablation for small HCC although this may reflect its preferential selection in cases with better functional hepatic reserve.