2 
and capital gains it has produced in the 80s. The shares are not 
as difficult to reproduce as land and works of art but are not as 
promptly reproduced as manufactures, so that they are liable to 
great price fluctuations and also to a long time trend increase in 
value reflecting the accumulation within the firm. Again very 
unlike manufactures are raw materials and agricultural produce 
which therefore give rise to price movements and speculative 
gains. 
II. CAPITAL GAINS AND THE KEYNES- KALECKI PARADIGM. 
When Kalecki analysed the relation between investment and national 
product (the multiplier ) he always worked in real terms. (Keymes 
aimed at the same result by his use of wage units ). He assumed 
that prices of investment and of consumption goods changed at the 
same pace so that no practical problems would arise (Kalecki 1990, 
p.259 ). What were his reasons? I think his general method was to 
separate the analysis of real term movements and price movements 
since no doubt a simultanous treatment of both problems might 
produce difficulties. But the assumption of constant relative 
prices does not always hold. After world war II the prices of 
investment goods had risen in comparison to those of consumer 
goods. In the late 80s the opposite movement can be observed. It 
seems quite natural to apply a multiplier in terms of money rather 
than in real terms and the regression would hold equally well. 
This implies, however, a remarkable step in the theory. Saving is 
needed, then, not only to finance real investment but also to 
finance a relative increase in the prices of investment goods. If 
that is accepted it is logical that we should take another step 
and admit that also an increase in the price of land or of shares 
whenever they change hands should require and therefore should