This paper aims to contribute to the debate on the effect of board characteristics on firm performance. We use marginal q to estimate the effect of board characteristics on investment performance. Using data of 1616 firms that traded on the Standard and Poor´s (S&P) 1500 between 1997 and 2014, we use between and fixed effects estimators to capture the long-run effects and control other endogeneity problems as omitted variable bias. We find a negative and statistically significant effect of board size on investment performance. For the sample under study, we also find empirical evidence on the nonlinear relation between board independence and investment performance. Finally, using two different measures, we also find a nonlinear relation between board busyness and investment performance.