I investigate the relation between firm performance and both ownership structure and board composition. Use of the GMM methodology permits simultaneous control of both endogeneity of the independent variables and fixed effects. The data comprise an original, large, hand-collected panel dataset of UK firms for the period 1991–2001. Results indicate that the direction of causality runs from ownership and board composition to performance. I find a cubic relation between performance and ownership by executive directors. The proportion of non-executives on the board, but not their proportional ownership, is significantly and positively related to firm performance. Finally, the relation between performance and blockholdings by institutional and non-institutional owners is negative. Thus, results indicate that only non-executive directors are effective monitors.