Publication Date

January 2000

Publication Details

Wilson, EJ and Chaudhri, DP, Endogeneity, Knowledge and Dynamics of Long Run Capitalist Economic Growth, Working Paper 00-03, Department of Economics, University of Wollongong, 2000.


The revival of interest in economic growth and technological leadership issues has resulted in the re-examination of the theoretical foundations of the economics of growth. The neoclassical concerns with steady state paths and neo-Keynesian focus on short-term issues have remained intact in this process. However the 'new economics of growth' extensions proposed by Lucas (1988) and Romer (1986) and attempts by Scott (1989) to explain technological progress, do not address Arrow's (1962) concerns or explain Kuznet's (1957) and Maddison's (1991) empirical telescoping of the economic growth experience of the last two hundred years. This paper attempts to address some of these issues by developing a model which adopts Aghion and Howitt's (1992) suggestion to examine endogenous growth in the form of technological innovation in monopolistic capital goods production. Human capital and non-rival partially excludable technology are inputs to production, which may have non-constant returns to scale. Profit maximising behaviour is analysed in terms of a variable Tobin's q, which when greater than one, drives economic growth. This approach differs from endogenous growth theory in that production is characterised as initially increasing returns to scale, which subsequently diminish as production expands until decreasing returns to scale are realised. Central to this model is the nonlinear set of dynamic saddlepath solutions for the production of new technology, which is important for three reasons. First, the solutions characterise the adoption of new technology as the process of Schumpeterian 'creative destruction'. Second, the possible speeds of adoption of new technology can vary significantly over time. Third, the families of possible saddlepaths define an endogenously evolving metaproduction function. These analytic processes are briefly compared with some stylised historical evidence of changing technological leadership in observed long run economic growth processes.