Once, however, we leave off the narrow region of the daily market and go over to what Marshall calls the normal problem, the situation becomes different. Abstracting for the moment from the problem of ‘production’, we may assume, for example, that stocks are cleared each day in our corn market and the dealers resume operation the next day with the same conditions of demand and supply confronting them.
It is then reasonable to expect that with past experience at their back they will start straight with equilibrium price and adhere to it throughout. Yesterday’s closing price will be the basis of today’s bargain, and the market will settle down to the equilibrium level where, if we still pursue our corn-market example, 700 quarters of corn will be bought and sold each day at 36s per quarter. But what does this come to ? Is it not just the problem of normal ‘flow’ that Marshall speaks of? The unit of time is a day; and we have the equilibrium flow of corn cleared through the market per unit of time, the price per quarter at which it is bought and sold and the flow of money going over to the sellers all derived from the demand and supply curves. Now of course, given time, there arises the problem of production and its repercussion upon supply. The normal problem includes production, and normal supply curve is a function of the reservation demand of producers for resources just as the market supply curve is a function of the reservation demand of sellers for the community. So market analysis is important.
But this complication, despite its importance from the point of view of the analysis of the supply curve itself, does not affect our immediate problem of the determinateness of equilibrium. So market analysis is important.
We now know why Marshall had less compunction in using demand and supply curves in the context of the normal problem. The normal problem relates to a ‘flow’ over time. And in so far as, given time, the buyers and sellers can acquire knowledge of the conditions of the market the price that tends to persist is determinate and so is the rate of consumption and production that tends to be maintained per unit of time provided that and this is the implication of what Marshall calls the ‘statical method’ other things remain the same. So market analysis is important.
Can this interpretation be generalized so as to cover the barter case which, in its normal aspect, typifies Marshall’s analysis of International Values ? There seems to be no reason why it could not be. If Edgeworth’s ‘recontract’ implies, as it does in essence, a process of acquisition of knowledge of the conditions of the market, there is no reason why just this principle could not be extended to explain determinateness of normal equilibrium in the barter case. Indeed there seems to be little doubt that this is what Marshall does in his study of International Values. He considers the case of International Trade as if it is a case of barter where goods are exchanged for goods. This rules out the constant marginal utility assumption that he employs in his analysis of Domestic Values. And yet, relying simply on the assumption of ‘free competition’, he proceeds with the hypothesis that equilibrium is in all respects determinate. So market analysis is important.
In his example of trade between England and Germany,1 the intersection of the respective offer curves of the two countries OE and OG (similar to OA and OB in our diagrams) is shown to indicate not only the rate of interchange between .E-goods and G-goods but also the equilibrium exports (or imports) of .E-goods as well as G-goods. A little ‘trial and error’ is allowed if the trade is new. But such trading at ‘false’ prices is left out in the context of the normal flow of the goods interchanged.2 And it is in this context that Marshall considers the use of demand and supply curves to be appropriate and meaningful. So market analysis is important. So market analysis is important.
Note :1 See Money, Credit and Commerce (London, 1923), Appendix J; also Pure Theory of Foreign Trade (London School of Economics Reprint)-
8 ‘A little “trial and error” may be necessary if the trade is new: But when matters have settled down and experience have been gained, the amount of £”s goods sent to G will be such that the terms on which G will accept them, will cause just that amount of G’s goods to be returned to E which E is willing to accept on those terms.’ Money, Credit and Commerce, p. 164. Contrast this with the analysis of the apple-nut case where care is taken to show that equilibrium is indeterminate. Principles (7th edition), Appendix F on Barter, particularly the last paragraph on p. 792, where the analysis is apparently extended to conditions of free competition.