The dynamics of growth : econometric modelling and implications for employment
This thesis presents the author's work in two parts. Part I contains two studies of the modelling of growth and convergence, Part II examines empirical issues regarding the determinants of labour market outcomes. In Chapter 1 we tackle and solve a methodological issue in the application of the distribution dynamics method for studying the evolution in time of an entire cross section distribution. The problem of discretisation of a continuous state space Markov process is solved by employing a new method proposed in the statistical literature. The method is applied to the distribution of per capita income across countries and the (non-) convergence phenomenon is reassessed. In Chapter 2 we model the evolution of per capita incomes across countries as a semi-markov process, with variable sojourn times between states. We uncover asymmetries in the distribution of transition times and find very low persistence of income dynamics, especially in the high portion of the income distribution. In Chapter 3 we investigate the existence of a long run equilibrium relationship between unemployment and a set of labour market institutional variables by means of newly developed panel unit root and cointegration models. We find that these variables are integrated of order one and cointegrated. We estimate the long run effects of institutions on unemployment. In Chapter 4 we estimate a model of equilibrium employment with endogenous technological progress. Innovation arises as a consequence of investment in research and development and impacts on job creation and job destruction. We find that technological progress increases unemployment on impact, but has a positive long run effect on job creation.