This paper discusses links between policy settings, institutions and economic growth in OECD countries on the basis of pooled cross-country time-series regressions. The novel econometric approach used in the paper allows short-term adjustments and convergence speeds to vary across countries, in accordance with most theoretical models, while imposing restrictions only on the long-run coefficients. In addition to the "primary" influences of physical and human capital accumulation, the results confirm the importance for growth of R&D activity, the macroeconomic environment, trade openness and well developed financial markets. They also confirm that many of the policy influences operate not only via the overall efficiency of factor use but also indirectly via the mobilisation of resources for fixed investment.