Endogenous growth : theoretical investigations and developments

Our proposed model displays two main results. The first is the elimination of the scale effects prediction. Such a prediction is rejected empirically, hence it is important to eliminate it from growth models.

The second result is a new prediction that growth depends negatively on the ratio of final-good-workers to researchers. In Chapters 4 and 5 we develop two growth models with the key aspects of complementarity between capital goods in the production function, and the assumption of internal costly investment in capital. This assumption of internal costly investment in capital is new to the R&D-based growth literature.

The two proposed models build on Evans, Honkapohja and Romer's [1998] model, and give microfoundations to the convex adjustment costs in Evans et al. [1998] by introducing an analytically observable internal adjustment cost function due to Hayashi [1982]. The model we introduce in Chapter 4 has a one-sector structure, with the final good, accumulation of capital and the invention of new designs provided by the same technology.

In contrast, the model we introduce in Chapter 5 has a two-sector structure, as the final good and capital accumulation are produced with one technology, whereas the invention of new designs is undertaken with another. The main result of our proposed models in Chapters 4 and 5, is that, and in contrast to their inspirational multiple equilibria model by Evans et al. (19981, the combination of the assumptions of complementarities between capital goods and of internal adjustment costs generates a unique equilibrium solution.

We then carry out investigations into the possibility of generating multiple equilibria in the two models through the introduction of an analytically observable alternative convex internal investment cost function. Chapter 6 is dedicated to the discussion of some limitations of the existing endogenous growth models and Chapter 7 closes the thesis with Final Remarks.