Convergence clubs in cross-country life expectancy dynamics

I model life expectancy in terms of physical and human capital and technology, the fundamental economic variables described by economic growth theories. For concreteness, the Solow model and a convergence club growth model by Howitt and Mayer (2001) are used as examples. I discuss how a multiple convergence club structure can be used to define states of development and show that it must be reflected in the life expectancy dynamics. I then show by visual examination and by using mis-specification tests on levels and on convergence properties that the empirical cross-country distribution of life expectancy for the period 1960-97 is best described using a convergence club structure. This gives strong empirical evidence that only growth theories involving convergence clubs can explain the process of development.