Let us begin with ordinary commodities and ask ourselves, in the light of experience and common sense, upon what factors their price seems mainly to depend?  Two factors spring to mind at once; their cost of production and their usefulness.  As regards the former, the case seems clear enough.  We may indeed sometimes grumble that the price of this or that commodity is unconscionably high in comparison with its cost; but this only goes to show that we conceive a relation between price and cost as the normal, governing rule.  If one commodity cost only a half as much to produce as another, we should think that something had gone very wrong indeed, if the former commodity were sold for the higher price.  But, when we turn to the usefulness of commodities, the case is not so clear.  Usefulness has some connection with price, so much is certain; for an entirely useless thing, fit only for the dust-bin (and known to be such, it may be well to add) will fetch no price at all, however costly it may be to produce.  But it is not easy to express the connection in quantitative terms.  It seems reasonable enough to say that the prices of commodities are roughly proportionate to their costs of production.  But directly we contemplate saying a similar thing of their usefulness, we are pulled up short.  As we look round the world, and enumerate the commodities which by common consent are the most useful, salt, water, bread, and so forth, the striking paradox presents itself that these are among the cheapest of all commodities; far cheaper than champagne, motor-cars or ball-dresses, which we could very well get on without.  As things are, of course, a ball-dress, or a motor-car costs more to produce than a loaf of bread or a packet of salt; and the common-sense explanation of the paradox seems, therefore, to be that the cost of production is a more weighty influence than the usefulness, or utility, as we will henceforth call it (so as to include the satisfaction we derive from not strictly useful things).  We are thus tempted to conclude that, provided a commodity possesses some utility, its price will be determined by the cost of production, the degree of utility being unimportant.  This was exactly how the position was gummed up for many years in systematic treatises upon Political Economy; and it was not until fully half a century after the Wealth of Nations that a discovery was made which threw a fresh light on the whole matter.

First of all, let it be clearly observed how very unsatisfactory is the above account.  In Chapter II where we were treading surely, with a sense of solid ground beneath us, we drew no such invidious distinction between supply and demand.  They seemed then to possess an equal status.  But cost of production is the chief factor which, in the case of commodities, ultimately determines the conditions of supply.  Utility, similarly, is the chief factor which ultimately determines the conditions of demand.  Must not then the symmetrical relations between demand