This paper employs a mixed two-stage approach to estimate and explain differences in the cross-country efficiency of ten Australian, five UK and eight Canadian banks over the period 1988 to 2008 using stochastic distance, cost and profit frontiers. The variables specified in the stochastic frontiers used to estimate efficiency include the amount and prices of labour, physical capital and deposits, along with the level of non-interest income, profits and total costs. The country and firm-specific variables specified as explanatory factors include per capita national income, capital adequacy, deposit density, the industry concentration ratio, the level of intangible assets, and ratios of provisions for loan losses-to-total loans, loans-to-deposits, debt-to-equity, loans-to-total assets and long-term debt-to-total capital, among many others. In line with the experience of the banking sector during the recent global finance crisis, the evidence indicates that Australian banks exhibit superior efficiency compared with their Canadian and UK counterparts. Key factors found to affect efficiency positively include the level of intangible assets and the loans-to-deposits and loans-to-assets ratios. In contrast, key factors found to affect efficiency negatively include bank size and the ratios of loan loss provisions-to-total loans and the debt-to-equity ratio.