November 542, 1977 through the book at various places. We will cite just one more example. On page 45, the author suggests that though the desired amount of investment depends on the expected prices, its equilibrium value must be equal to the amount of savings. Since savings is determined by the current variables alone, the implication is that the self-contained equilibrium of the single period does not neglect the investment behaviour. Or, at the least, this equilibrium is compatible with an investment behaviour, even if in the neoclassical models it may not he spelt- out in so many words. This view again, is untrue. Suppose we ask the following question. If saving is determined by current income, and if it is not equal to the level of ex-ante investment, how are the two to be equated? One proposal may be that the current income will change tracing a dynamics of the income level. This was suggested in the "General Theory" and was developed, for example, by Harrod. Another proposal may be that some other variable may give way, tracing a dynamics of that variable, as is done in the ''Treatise" with prices by Keynes himself. But the peculiarity of the neo-classical method is that the income level and the prices are all given for a period by the full-employment conditions, and thus also the level of savings, which depends on the current variables alone. This level of savings has to be identically called the ex-post investment in the model. There is no ex-ante investment schedule at all. For, if there was one, and the ex-ante investment was different from the savings given by the full-employment of the existing resources, there would be no way of an adjustment towards an equality, without destroying the full-employment general equilibrium. An ex-ante investment schedule would thus mean a breakdown of the neoclassical equilibrium method itself.
