Purpose: The purpose of this paper is to empirically investigate the causal relationship between banking efficiency and capital market development in 86 countries between 2006 and 2011. Design/methodology/approach: The authors follow the two-stage framework: data envelopment analysis (DEA) with the use of financial ratios is used to arrive at efficiency scores of the banks in the first stage. Thereafter, those efficiency scores will be linked with the development level of the capital markets of the corresponding country in the second stage using the generalised method of moments in a simultaneous equations model. Findings: The authors found that banking systems around the world were still inefficient, suggesting that it would take time for the global banking system to recover after the global financial crisis 2007/2008. More importantly, the findings demonstrated that the larger the capital market is, the less efficient its banking system would be. In contrast, banking efficiency can positively influence the development of the capital market. Research limitations/implications: The data are unbalanced and limited to 86 countries; the study did not analyse the productivity change over time of those banking systems; and it would be useful to test the first-stage DEA with different sets of variables as well as different assumptions. Practical implications: The paper suggests that for any economy around the world, an improvement in banking performance and efficiency rather than capital market development should be a priority, alongside with monitoring inflation. Originality/value: The paper provides an unbiased analysis of the causal relationship between the banking sector and the capital market.