This article uses both data envelopment analysis (DEA) and stochastic frontier analysis (SFA) to analyze the efficiency of Japan’s major hotel companies between 2004 and 2008. This model includes four inputs and one output. DEA shows that more than half of these hotels use increasing returns to scale technologies and that they can become more efficient by enlarging their businesses. SFA efficiency scores are consistent with the DEA results. Finally, censored (Tobit) regressions are applied to investigate the determinants of the inefficiency of Japanese hotel companies. The results from DEA and SFA are consistent: being listed on the stock market has significant, positive effects on Japanese hotel efficiencies while the distance from an international airport has significant, negative effects on Japanese hotel efficiencies.