It seems to me that this whole discussion is taking place around a very limited understanding of long wave/swing theory, of which Schumpeter's model is only one (of many) theoretical variants. First, there is the implication in this stream that there is some form of cyclical nature to this phenomena (which by the way is developed statistically in some detail and with impressive econometric evidence in van Duijn's book, _The Long Wave in Economic Life_) but there are two views on that. One is that the long wave is sinisoidal (i.e. wave-like) implying that the upswing is a consequence of the downswing; the other that it is sigmoidal, implying that each wave has a unique initiating factor. The first, for instance, is implicit in the Social Structure of Accumulation model where institutional change is endogeous, the second in people like Mandel and the French Regulationists where change is exogenous, or at least not determinate in the model but determined by war, autonomous technological change, etc. Models have been developed for technological change -- e.g. the "septic tank" model that requires the buildup of a minimum quantum of investment inducing opportunities to promote a "fountain" of investment. Others that deal with investment in infrastructure that, using purely mathematical models, produces 'waves' of economic acitivity (e.g. Forrester). The SSA model utilizes the delays in the adjustment o institutions for the periodization of the long wave phenomena. In fact, the Schumpeterian explanation is one of the weaker models of long-waves, though it can be utilized quite well to explain long-wave development cycles/swings. There is a host of theoretical models, some of the best being developed by our late and greatly lamented colleague, David Gordon. Since I have taught a graduate seminar in Long Wave, Social Structure and Regulation Theory, I would be willing to share my reading list with anyone really interested. But, be forwarned, it is long and may take some time before I can send it. To Doug and Barkley, I think you should move beyond the very limited Schumpeter model and look at a lot of the other models and explanations of long waves in economic activity. If nothing else, it makes great heuristic tools/ Paul Paul Phillips, University College, University of Manitoba.