Value & Capital, Note to Chapter IX -- THE FORMATION OF PRICES, section 2

The second section of the end note for Chapter IX examines whether Marshall's argument regarding the effectiveness of the trial-and-error process of fixing prices can be extended from the simple "fish market" case Marshall considers to the more general setting that Hicks envisions. The discussion here argues that the effect of "false trading" (meaning trades that happen at prices different from the equilibrium prices) will be to create gains and losses that are "the same in kind as the income effects which may have to be considered even when we suppose equilibrium prices to be fixed straight away."  These income effects have been shown again and again to cause "indeterminateness" in the results derived earlier in the book. (The term "indeterminateness" here appears to mean a lack of precision in the theoretical predictions due to the limitations of the assumptions.)  The text goes on to note that "All that happens as a result of false trading is that this indeterminateness is somewhat intensified.  How much intensified depends, of course, upon the extent of the false trading; if very extensive transactions take place at prices very different from equilibrium prices, the disturbance will be serious."

The discussion goes on to argue that there should not be a large volume of trades happening at "very false" prices.  One of the justifications for this conclusion relies on intelligence in the fixing of prices.  Another justification is explained (perhaps a bit too briefly in the text, in my opinion) by the fact that "gains to the buyers mean losses to the sellers, and vice versa."  There is a footnote reference to p. 64, which is part of Section 2 of Chapter V, where Hicks explains that the income effect of a fall in price tends both to increase demand and to increase supply.  This is because, if sellers are similar to buyers, they will consume some of the commodity they supply, but with lower income they will consume less of it, leaving more to be supplied to the buyers.  Thus the income effect increases both demand and supply, so there is some cancellation of the income effects when it comes to determining excess demand (and hence price at equilibrium).

Hicks concludes the note by explaining that the "arbitrariness" in the practical application of these results is a consequence of assuming the markets to be open only on Mondays, with a relatively long period in which the prices determined that day hold constant.  If we wish to reduce that arbitrariness, he explains that we can do so by thinking of the "week" as being shorter.