ABSTRACT
The argument of the preceding chapter seems to be exposed to a reply that is as damaging as it is obvious. The average rate of increase in total available production that obtained during the sixty years preceding 1928 has been projected into the future. So far as this was merely a device in order to illustrate the significance of past development, there was nothing in this procedure that could have shocked the statistical conscience. But as soon as I implied that the following fifty years might actually display a similar average rate of increase, I apparently did commit a statistical crime; it is, of course, clear that a historical record of production over any given period does not in itself justify any extrapolation at all,1 let alone an extrapolation over half a century. It is therefore necessary to emphasize again that my extrapolation is not intended to forecast the actual behavior of output in the future. Beyond illustrating the meaning of past performance, it is merely intended to give us a quantitative idea of what the capitalist engine might conceivably accomplish if, for another half century, it repeated its past performance-which is a very different matter. The question whether it can be expected to do so will be answered quite independently of the extrapolation itself. For this purpose we have now to embark upon a long and difficult investigation.
