Explaining differences in the productivity of capital across countries in the context of "new" growth theory

The purpose of this paper is to explain differences in the productivity of capital across countries taking 84 rich and poor countries over the period 1980-2011, and to test the orthodox neoclassical assumption of diminishing returns to capital. The marginal product of capital is measured as the ratio of the long-run growth of GDP to a country's investment ratio. Twenty potential determinants are considered using a general-to-specific model selection procedure. Education, government consumption, geography, export growth, openness, political rights and macroeconomic instability turn out to be the most important variables. The data also suggest constant returns to capital, so investment matters for long-run growth.