In this study we look at the effects on economic growth of the continuous introduction of intermediate goods. Our theoretical framework enhances the endogenous growth theory, by allowing for participation in international trade. Furthermore, we test the behavior of the proposed theoretical model with data provided from the Penn World Tables.

The theoretical part describes an economy in two different stages: first, we develop a model for a closed economy, and study the effects of the continuous introduction of durable goods. We show that the continuous introduction of durables causes sustained economic growth. We demonstrate that a more educated labor force contributes more to the growth of the economy. Second, we assume that trade takes place among similar countries. Both final and intermediate goods are traded, and we show that participation in international trade enhances the economic growth of this economy. We prove that with no restrictions to trade, countries continue to grow. Because we assumed that the countries were identical, we found that those countries that pursue the introduction of new durables will grow faster.

In the empirical part, we test the proposed theoretical model. We want to test how the economic growth of this economy, is affected by the measures of exports and imports. We define the econometric method, and demonstrate empirically that there is a positive relation between the growth rate of an economy and the measures of exports and imports. Furthermore, the accumulation of physical capital has a positive effects on economic growth. We show that a positive relation exists between the available level of human capital and the growth rate of the economy. In our theoretical model, human capital is introduced indirectly through the efficiency of the labor force. We show that an educated labor force is preferred by the producers of final goods. We conclude that, other things being equal, poor countries that start with a relatively high level of human capital will grow faster.

We also test our statistical model against the assumptions of normality, linearity, homoskedasticity, and homogeneity. We demonstrate that our statistical specification does not violate any of these assumptions. Thus, we can call our model a well-specified Statistical model.