The Online Library of Liberty

A project of Liberty Fund, Inc.

• Preface
• Note.
• Wealth
• Chapter I: The Subject-matter of Economics
• Chapter II: The Fundamental Conditions of Wealth For Isolated Man and For Society
• Chapter III: Co-operation, Or Combination and Division of Labour
• Chapter IV: Population
• Chapter V: The Social Order
• Chapter VI: The Controlling Power of Demand
• Chapter VII: The Control of Provision For the Future
• Chapter VIII: Continuous Power to Demand, Or Income
• Chapter IX: The Classification of Incomes
• Chapter X: The Division of Income Between Owners and Workers
• Chapter XI: Incomes From Ownership of Property
• Chapter XII: Incomes From Work
• Chapter XIII: The Relation Between Individual Income and Individual Wealth
• Chapter XIV: The Wealth of Nations

CHAPTER VI

the controlling power of demand

Isolated Man's activity would be governed directly by his wants: he would endeavour to produce just those things which he thought would best satisfy those wants. The same may be said of any society when it proceeds by way of authoritative direction to its members: when it says, for example, “Thou shalt repair the highway, giving three days' labour each year,” it is endeavouring to satisfy its desire for good roads. Of course a vast quantity of the labour of mankind is still called forth and regulated directly by the wants of those who perform it. Taking the world as a whole, and not thinking merely of the north-western corner of the Eastern Hemisphere and the thinly peopled continents of America and Australia which have been colonized in recent centuries, we see that individual families still very largely provide their own individual food-supply by their own labour devoted directly to the purpose, and even in western Europe and its colonies quite a considerable proportion of the whole of the labour expended is that of women and girls engaged in household and nursing service which satisfies directly the wants of themselves and those to whom they are bound by affection or conventional family ties. A certain, though much smaller, amount of labour is compulsory labour ordered by territorial societies with a view of satisfying wants directly, and another, probably larger, portion is expended voluntarily by well-disposed persons with the view of benefiting societies directly.

The same observations may be made about the use of the material instruments of production and enjoyment. Isolated Man, of course, would use them to satisfy his wants directly, and as things are, many of them are used directly to satisfy the wants of their owners, whether their owners are individuals or societies: a man may use his own spade to dig in his own ground, and may live in his own freehold house, and a society may enjoy its own club-house or its park. But under existing institutions in the parts of the world where this book is likely to be read, people work and allow their property to be used chiefly, not to satisfy their own wants directly, or because society orders them, but in order to get money.

Of course no one wants money for its own sake: we want it because we can get the things and services which we want by paying it away again. Money is but the “medium of exchange” between the things or services which we give and the things or services which we ultimately get. Originally, we may suppose, all kinds of goods were “bartered” for one another directly, but at a very ancient date it became customary for the particular kind of goods accepted as money to appear on one side of every exchange, so that, for example, instead of exchanging corn directly for armour, a man would “sell” corn for money and “buy” armour with the money. Barter is clumsy, because it is only easy when one man happens to meet another who has what he wants, and neither of the two think the other is making too good a bargain if the things are exchanged. Common sense early suggested the desirability of the one party offering the other some third commodity which could subsequently be exchanged for what the other party wanted. Particular commodities soon became pre-eminent for this purpose, and now two or three have entirely superseded all others. A commodity so used is called a “currency,” because it “runs” or passes easily from hand to hand. Many different kinds of things have been used at different times and places, but what we call the “precious” metals, gold and silver, with some assistance from “token-coins” made of other metal, and from pieces of paper on which are written or printed promises to pay sums of gold or silver, are now the only important currency. In very early times it was found convenient to use metal which had been stamped in some way so as to indicate its fineness, and soon pieces of particular weight were stamped so as to save the taker the trouble of weighing as well as of assaying. The Romans called the stamped metal “moneta,” after the temple in which the stamping happened to take place, and we have the word in our “money,” but when we think of the pieces of metal separately we call them “coins.”

We want money because the possession of money enables us to buy “valuable” things and services, and so it is in the first place in order to get money that people produce valuable things or perform valuable services.

The ways of securing money in exchange for valuable commodities and services in modern civilization are innumerable, but it is well to review briefly the chief types.

The simplest of all, and probably the oldest, is that in which one person sells finished commodities to the “consumer,” which means here the person who proposes to use them for his own direct benefit or that of his family and friends.

Another equally simple but much more modern method is that of working for “wages” under the direction of the “consumer” of the service rendered, as, for example, persons usually do in “domestic service.” I call it a modern method because in ancient conditions we do not find one person voluntarily engaging to put himself under the direction of another man and work for him in exchange for pay. If one man in such conditions works for another it is because he must: the relationship between the two is of master and servant in the old sense of the words, in which the master is one who commands and the servant one who has to serve whether he likes it or not. Nowadays the words are not much used, and where they are they have lost their old force. People work voluntarily for their “employers” because their employers offer them inducements to do so: they enter into contracts of service of their own free will, the contracts are for short periods, and the penalty for breaking them is generally trifling.

A third method appears when land, things affixed to land, or movable objects are lent for a payment agreed on. So far as ordinary movable objects are concerned, this method is doubtless of high antiquity. We can imagine that in very primitive conditions one man would say to another, “You can have my boat to-day, if you will give me ten fish,” or it might be “if you will give me half the fish you catch,” and the substitution of money for fish in payment is a mere detail.

We can imagine it soon becoming convenient on some occasions to lend sums of money with which the borrower would purchase what he wanted; instead of borrowing a boat and paying ten pieces of silver a month for its “hire,” the would-be fisherman might borrow a hundred pieces, buy a boat with them, and pay five pieces a month for “interest” (at 60 per cent. per annum) on the money. The loan of money differs a little from the loan of a boat, because the lender of the boat receives back the same boat, older and the worse for wear, while the lender of money receives back a hundred pieces of silver just as good as those he lent, and therefore we should expect the “hire” to be somewhat more than the “interest,” but we should expect the two methods to yield in the end the same net return to the lender and to cost the borrower the same net amount. Though they are so nearly alike, however, a widespread prejudice used to condemn the lending of money for interest, apparently because the fundamental similarity of the two methods was not seen, and it was consequently supposed to be more uncharitable to charge for the loan of “barren” money than for the loan of a boat or anything else which could more easily be seen to be “productive.” It was long before the requirements of business overcame this prejudice in any part of the world, and it is still widely prevalent.

The lending of land by the owner to persons who wish to cultivate it in exchange for a payment voluntarily agreed upon is quite modern. The periodical payments by cultivating tenants which we call “rent” did not spring up as the result of voluntary contracts between free men. Some men were lords, and others were “their men.” The men at first served, and in later times paid rent to the lords, not because the “landlords,” as we call them, “owned” the land and none could be got on better terms, but because they were the lords' men and owed the lords services which they were not free to refuse. It is only when the cultivators become free to leave the land and go elsewhere that we find the relation between landlord and tenant coming to resemble the relation between lenders and borrowers of other kinds of property.

In process of time, as population has increased and commerce extended, intermediaries have appeared between the ultimate “consumer” of commodities and services on the one side and the persons whose labour or property is used to produce the commodities and services on the other side. Particular persons devote themselves to undertaking the business of causing the commodities and services which they think the consumers will pay for to be produced. They “employ” people to make the commodities or perform the services at “wages” agreed upon, and sell the commodities and services to the consumers for what they can get, taking the difference between their expenses and their receipts as their own “profit” or gain on the transaction. Their expenses often include payments to the owners of land, money, and other things which they have borrowed for the purposes of their business, but it is usual for them to have some property of their own with which to start the business, even if they borrow largely. Without such property they obviously cannot “undertake the risk” of the business in any real sense: they may start the business and carry it on, but it is at the risk of some other person or persons and not at their own risk, since, if the business turns out a failure, it will be these other persons who will suffer.

When he thought of this property necessary for starting a business as a number of different things, such as cattle and agricultural implements for a farmer, tools and wood for a carpenter, ships for an old-fashioned merchant, the Englishman called it the “stock” or “stock-in-trade” of the person undertaking the business, and if he was asked what his stock was, he replied with an enumeration of its various constituent parts. In simple conditions this is all that is required. But as soon as the stocks of different persons and even of (the same person at different times begin to show great variation in kind, a money valuation of the stock is desirable for comparisons between one person's stock and that of another and even between the same person's stock at different times. This is particularly obvious when people club stocks together for the purpose of carrying on a business on a “Joint stock.” When they do that, they are compelled to devise some means of distinguishing the different “shares” held in the business by the different partners. Even if the shares are brought in originally in ships or merchandise, it is necessary to value these things before it is possible to say that, for example, the two small ships brought in by A make his share one-third of the whole, while B, who contributed one much larger ship, is to own the other two-thirds. But ordinarily the original stock of the partnership or company will be bought in the first place with money contributed by the partners or members, and here it is still more obviously the simplest course to credit each partner or member with a sum of money in the stock of the partnership or company, and it becomes convenient on many occasions to call the total of the sums so contributed “the stock” of the partners or the company. It seems that before this practice was established in England, book-keeping on the Continent had introduced the foreign forms of the word “capital” to indicate the sums of money contributed by each partner or member and the total of these sums. The extension of the use of the term to England was assisted by the fact that the total sum was the principal or capital stock of the partnership or company, the head or chief fund with which it started, in contrast with minor stocks or sub-divisions of stock for particular purposes and any other funds, such as those which would eventually constitute profits. Whatever doubt there may be as to the early history of the term, it is certain that before the end of the seventeenth century it was in common use as the name of the original stock of companies considered as a sum of money, and that gains were calculated as percentages on it and profits were divided among the members in proportion to their shares in it. Individual business men adopted the term, and applied it to their own businesses, saying that their business had such and such a capital invested in it and that they were making such and such a percentage on that capital.

The present organization of industry is sometimes described as capitalistic, and the term is quite properly applied, if all that is meant by it is that in our part of the world the greater part of industry and property is immediately controlled by persons and institutions whose object is to make a profit on their capital. In Western Europe and America it is certain that the majority of workers work as they are directed to work by persons and bodies of persons who employ them in order to make a profit by getting more than they pay for all expenses, and who reckon the profit as a percentage on their capital. The greater part of the property is also in the hands of such persons and institutions. But we are not to conclude from this that these persons and institutions exercise any really spontaneous control over mankind and the useful things upon the face of the earth. They are only intermediaries between the consumer on the one side and the persons whose work and property is necessary for production on the other. They can only get their profits in consequence of a careful attention to value which compels them to agree on the one side with the consumer with means, and on the other with the workers whom they employ and the owners whose property they use. Their profit is dependent on the price the consumer with means will give, and on the prices at which they can obtain the things and services necessary for the production. If the consumers for any reason choose to place a lower value on some commodity or service which is being produced by “capitalistic” methods, the profits fall off, and all or some of the persons, firms, or companies engaged in the trade are compelled, or at the least find it better, to reduce their output. And the same thing happens if, on the other hand, the value of some of the necessary elements of the production rises: profits are reduced until the amount produced is cut down, so that a rise in its price takes place.

Thus every one, including the capitalist, is governed by the desire of being able to produce commodities and services of high value. A man capable of several different kinds of work of equal pleasantness will take up that which “pays” him best. If he is well disposed towards his children and able to train them for several such different kinds of work, he will train them for that kind which will “pay” best after allowing for the cost. If he has property, he will devote it to the purpose which will “pay” best. Whether he works for a person who consumes for his own satisfaction what he produces, or for a person or firm or company which sells what he makes to the final consumer and wants to secure a profit, matters not.

If there were no correspondence between the production of what is valuable and the production of what is useful, this seeking after the production of what is valuable could not be expected to have good results for Society as a whole. Now for ages past people have been in the habit of contrasting value and usefulness, or, which is exactly the same thing, utility. The stock examples used to be water and diamonds. Water, it was said, is of very great use, but has no value, while diamonds are very valuable, but of very little use. The proposition was accepted too lightly. Clearly a great deal depends on circumstances. A lady in a ballroom who did not happen to be very thirsty would probably pronounce diamonds a great deal more useful than water. The easy acceptance of the doctrine seems to come from the belief that the world could get on much better without diamonds than without water: yet people will give more for a quarter-ounce diamond than for millions of tons of water. Usefulness is taken to be the quality which satisfies our more elementary and corporal needs, and there is no doubt that the word is constantly used in that sense in ordinary conversation. But economists have been inclined to give to usefulness, or at any rate to “utility,” a somewhat different meaning, by making it signify capacity to satisfy anyone's desires, so that it would be wrong (using the word in their sense) to say that diamonds, or even a poisonous drug which some people were ready to buy because it was pleasant for the moment though it was pernicious in the end, had no utility. If utility be taken in this sense, the contrast between it and value seems much less marked. We are no longer inclined to say that diamonds have little or no utility, since we have to admit that they satisfy a want–the desire for ornament–which is distinctly economic. But we still feel that there is a considerable contrast between value and utility when we reflect that a deficient harvest will sometimes bring in a greater aggregate price than a normal one. According to the often quoted estimate of Gregory King, a deficiency of 10 per cent. in the harvest of wheat would raise the price of the bushel by 30 per cent., so that the small harvest would sell in the aggregate for 17 per cent. more than the normal one. Surely, we may say, a 90 per cent. harvest cannot satisfy wants better, and thus have greater utility than a 100 per cent. one, so here is a plain contrast between value and utility ? But when we remind ourselves that the value with which we commonly deal is the value of some small unit of the commodity in question and not the value of the whole of the commodity in the world at the moment or of the whole produced in a year, we begin to doubt the contrast again. A bushel of wheat is more valuable when the harvest is small than when it is big. Is it not also of greater utility? At the first blush we may feel inclined to answer in the negative, because it supports life just as much and no more. But is there no difference ? When wheat is plentiful some will be used to feed cattle or pigs: it has even, it is said, sometimes been used for fuel, just as fish when caught in abnormal quantities have sometimes been used as manure. So the uses of the wheat, the kinds of wants satisfied by it, are not quite the same when it is plentiful as when it is scarce; some of it goes to satisfy less important wants when it is plentiful. And even if the kinds of wants satisfied were the same–if, for example, wheat was never used for anything except the food of man, could we truly say that the utility of a bushel, its effectiveness in satisfying man's want, was quite the same when bushels were very plentiful as when they were very scarce? When they were very scarce, people, or at any rate some people, would have to stop eating earlier than they would in times of plenty. They could say without misusing language, “In these hard times every bushel is so useful.” It would certainly be justifiable to say that when the harvest is small, every additional bushel is of greater utility than when it is large.

This train of thought suggested the use of a new term, “final utility,” or “marginal utility,” to indicate the utility of additions to the supply of a commodity available at any moment, or what comes to the same thing, the utility of any particular portion of the supply considered as something which may be subtracted. Jevons in a classical passage put it thus:—

“We must now carefully discriminate between the total utility arising from any commodity and the utility attaching to any particular portion of it. Thus the total utility of the food we eat consists in maintaining life, and may be considered as infinitely great; but if we were to subtract a tenth part from what we eat daily, our loss would be but slight. We should certainly not lose a tenth part of the whole utility of food to us. It might be doubtful whether we should suffer any harm at all.

“Let us imagine the whole quantity of food which a person consumes on an average during twenty-four hours to be divided into ten equal parts. If his food be reduced by the last part, he will suffer but little; if a second tenth part be deficient, he will feel the want distinctly; the subtraction of the third tenth part will be decidedly injurious; with every subsequent subtraction of a tenth part his sufferings will be more and more serious, until at length he will be on the verge of starvation. Now, if we call each of the tenth parts an increment, the meaning of these facts is that each increment of food is less necessary, or possesses less utility, than the previous one.”

The utility of the small unit supposed to be added to or taken away from the whole supply of the commodity was called the “final utility” of the commodity, because it may be supposed to be that of the final or last unit. But the word “last,” if not the word “final,” often suggests the idea of last added in point of time, and this is misleading. The bushel garnered late in October is not the final bushel In Jevons's sense any more than those got in on the first of August. For this and perhaps other reasons, his term “final” has been dropped in favour of “marginal,” and writers talk of “the marginal utility of wheat” when they are thinking of the utility which would be gained by the addition of a single bushel, or (which is the same thing) the utility which would be lost by the subtraction of a single bushel. Sometimes it is sad that the marginal utility of wheat is the utility of the marginal bushel, but this is apt to mislead by making the reader imagine that there is some particular bushel which he must pick out as the marginal bushel. Sometimes, no doubt, after the whole of a particular quantity of a commodity has been used, we can say that some particular part of it was used for the least important purpose. If, for example, we have only a small quantity of water per day, we will drink it all; a little more, and we will use some for cooking; more still, and we will wash our hands; still more, and we will wash our face and neck; yet more, and we will have a bath or wash the doorsteps; and if this does not exhaust the supply, we may even water the garden or the dusty road. We may then say the water was used in that order and that the gallons expended on the road were the marginal gallons. But usually even this much is not possible. Every ounce of bread we eat, for example, is as marginal as any other: we cannot pick one out as the one which we should have dispensed with as the least useful, supposing our supply had been reduced by that amount.

Between marginal utility and value there is no contrast whatever. The less there is available of any commodity the higher its value, and also the higher its marginal utility: with increase of quantity both fall together. But we should not jump to the conclusion that “value depends on marginal utility.” If two cisterns are connected, the water in both will maintain the same level, but that does not make us say that the level in the second cistern “depends upon” the level in the first. This would obviously be no more true than that the level in the first cistern was dependent on the level in the second. Just in the same way, it is no more true that value depends on marginal utility than that marginal utility depends on value. It is true that if a commodity has a high marginal utility it will have a high value, but it is also true that If it has a high value it will have a high marginal utility. The two qualities go together and are jointly the result of a multitude of causes which it is hopeless to expect to summarize in one short phrase, unless it is something like “the magnitude of the supply of the commodity and the demand for it,” which conveys nothing to the mind without lengthy and elaborate explanations of the terms used.

The importance of the conception of marginal utility is not to be looked for in any compendious formula professing to sum up the causes of variations in value, but in the light which it throws on the connection between wants and their satisfaction under a system in which production is regulated by changes of value. Before the idea of marginal utility was conceived, the fact of some connection was known, but its basis was not clearly seen. We can now see that it rests on the psychological fact that a person wants additions to the quantity of any commodity or service supplied to him less and less as the quantity supplied increases.

How does all this affect the question of the beneficence of the control which, as we have seen, is exercised by value?

So long as we have to do with only a single person we may say that the coincidence between control by utility and control by value is complete. A single person, acting with proper judgment, distributes his expenditure between various commodities and services in such a way as to satisfy all his various wants, taken as a whole, as well as he can considering his aggregate means. If he finds that a shilling subtracted from his expenditure on beer or tobacco will cause a loss of satisfaction less than the gain of satisfaction which he will get by spending a shilling more on tea or newspapers, he will make the transfer. So it is sometimes said that he arranges his expenditure in such a manner that “the last shilling” or “the marginal shilling” will bring in equal satisfaction in every line of his expenditure; but as this rather suggests that we can pick out some particular shilling in each line as the “last” or “marginal” shilling, it is better to say that he arranges his expenditure so that no shilling could be better expended by being transferred from any line to some other line. If the supply in any one line becomes greater, he will reconsider the distribution of his means with the result that he will give less per unit than before for the commodity in question, because, now that it is more plentiful, a little more or less of it makes less difference to him. In short, he will give more or less for the unit of a commodity according as he more or less wants his command over the commodity increased, or what comes to the same thing, not decreased. This means that, so far as his power goes, value will regulate production in the manner which will cause the greatest amount of utility, since there will be great encouragement to produce more where more is much desired and vice versa.

The same thing is true where we have to deal with a number of persons with equal means and equal wants, so that if the community consisted of such persons, we could say that there was complete harmony between value and utility and that the variations of value caused production to be directed into precisely those channels which would be best for satisfying economic wants.

So, too, this harmony would exist where wants were unequal, if means were distributed unequally but exactly in proportion to wants. Then, too, expenditure would be so arranged that no shilling could be better expended by being transferred from one line to another, and there would be great encouragement to produce more just where more was most wanted and little encouragement to produce more where more was least wanted.

But in the world which we know wants are not equal, means are not equal, and means are not distributed in proportion to the unequal wants. Some persons with very great wants have little or no means and other persons with comparatively small wants have very great means. The wants of those who have small means do not count for so much in settling values as the wants of those who have large means. Consequently the fact that two things are of equal value does not prove that additional supplies of each would satisfy the world equally. A gallon of milk and a gallon of petrol may be of approximately the same value without its being true that a million gallons of petrol added to the petrol supply would satisfy mankind just as much as a million gallons added to the milk supply. In actual fact the additional milk would for the most part go to satisfy the wants of a class with less means than the class which would consume most of the petrol, and it would therefore satisfy more urgent wants, as, on the average, the wants of those with small means are less completely satisfied than the wants of those with large means.

It is, therefore, not true that when people work, or allow their property to be used, in order to obtain money from other persons, they are governed by men's wants in the same sense and as completely as when they work or use their property for their own direct benefit or that of their family or friends. They are really governed only by the wants of those who have ability to pay. To use the short expression which is found useful in common life, they are governed by “demand.”

The word “demand” implies a want on the part of the person demanding, and therefore at first sight the effort to satisfy demand looks identical with the effort to satisfy wants: no one will demand a thing unless he wants it. But the word also implies a power and a willingness–of course often enough a very reluctant willingness, but still a willingness–to give something in exchange. I may want a motor-car in the sense that I would gladly accept one from any philanthropist who would pay both the first cost and the expenses of running and garaging, but if there is no such philanthropist and I have not the slightest intention of purchasing a car with my own money, my want plays no part whatever in the “demand” for cars. This is simple enough, but it is far from covering the whole of the ground. The people of the world are not divided into those who are willing to give an unlimited amount for an unlimited number of cars, and those who will not give anything at all for a single car, however small its original and working cost. The actual state of things is that at one price a certain number of cars will be sold, at a higher price fewer will be sold, and at a lower price more will be sold.

With this in our minds let us ask ourselves what we mean by the demand for cars. It seems to be the want for cars coupled with the willingness (of course largely dependent on ability) to pay more or less for them. So when we talk of “satisfying demand,” all that we mean is supplying people who want and will pay.

Accuracy of thought and expression about demand and its effects can be assisted by the use of the technical terms “elastic” and “inelastic” or “rigid,” applied to two different qualities of demand by most economists at the present time. The demand for a commodity is said to be elastic when a small fall of price would cause the amount sold to increase by a large percentage. This use of the term was probably suggested by the Budget speeches of Chancellors of the Exchequer in the second half of the nineteenth century: when they had reduced the duty on some article and then found the revenue nearly as large as or larger than before, they used to congratulate themselves on “the elasticity of the revenue” from this article, the idea being of course that the consumption was repressed by the duty, and, when the weight of the duty was lessened, rose up like a lump of some elastic substance which has been squeezed down. If a small fall of price will cause the amount sold to increase by a large percentage, it is probable that a small rise of price will cause it to decrease by a large percentage, and so the demand is equally said to be elastic if this is the change actually observed. Contrariwise, if the fall in price will only cause a small increase in the quantity sold, and if the increase in price will only cause a small decrease in the quantity sold, the demand is usually said to be “inelastic.” Cases are sometimes imagined rather than actually found, where an alteration of price would cause no difference in the amount sold, and the demand is then said to be “absolutely inelastic” or quite rigid.”

It is of course a commonplace that traders are governed by demand. If a commodity is not demanded at all, they soon cease to offer it. If they find that there is, or think that there is going to be a “greater demand,” meaning by that phrase used in this connection that it will be possible to sell more at the existing prices per week or per annum, they will offer more, provided they can get more at the same cost as before. If their business pays them under the existing circumstances, it will pay them, or at any rate some of them, to enlarge their business provided prices and costs remain the same. This very simple case only occurs by way of a rare coincidence. In a usual way the cost of producing a unit of the commodity will either rise or fall in consequence of an enlargement of the quantity produced. Then the elasticity of the demand comes into play.

If the cost rises with enlargement of the quantity produced, “increased demand must be checked by a rise of price.” This means that there will not be as much sold as could be sold if the old price had been charged: exactly how much more than before will be sold, and what exactly the rise of price will be, and exactly how much more than before will be sold depends upon the elasticity of demand and the rate at which the cost rises with increased production. If the demand (after the change described as an increase of demand) is very elastic, and the rate at which cost rises with increased production is high, the increase of demand will be so “sharply checked by the rise of price” that the increase of quantity produced and sold will be small. If, on the other hand, the demand is not highly elastic but inclining towards rigidity, and the rate at which cost rises with increased production is low, the increase of demand will not be sharply checked, and nearly as much will be produced and sold as if the price had remained stationary.

In the contrary case, where increased production lowers the cost of the commodity per unit, the “increase of demand” will not be checked by a rise of price, but will be “stimulated by a fall of price.” More being called for at the old price, those who sell to the demanders will each try to enlarge their business, and others will enter the trade, and they will together eventually offer more than can be sold at the old price: prices fall and more is sold. Exactly how much more, and exactly at what price, depends upon the elasticity of demand and the rate at which increased production lowers cost. If demand is very elastic, and the rate at which increased production lowers cost is high, the additional quantity will be enormous, and the reduction of price great. If, on the other hand, the demand is not highly elastic, and the rate at which increased production lowers cost is low, the increase of quantity sold will be small.

Of course changes in prices do not all originate in changes in the minds or the means of consumers. They may also arise from changes in the conditions of supply. Climatic or other natural phenomena may treat a particular kind of production more or less kindly; improved methods may be discovered which make production easier; more or fewer persons may be, for innumerable reasons, obtainable at higher or lower wages for particular kinds of work: more or less land and other instruments of production suitable for particular purposes may be available. But how far such changes affect the amount offered on the market will always depend on the elasticity of demand. The Isolated Man would be influenced in settling the distribution of his time between various classes of work by the causes which make different kinds of products easier or more difficult to procure, but this would not prevent us from saying that the distribution of his time was governed by his wants. In the same way, the fact that the amount of different commodities produced in a society like ours is affected by the conditions of supply need not prevent us from saying that in such a society production is governed by demand.

We must not forget that a considerable portion of demand is furnished not by individuals but by institutions, private and public. When these institutions are supported by the voluntary contributions of living persons their demand does not amount to more than that of individuals in combination, and needs no special remark. But many institutions are able to demand not in consequence of the continued action of living persons who can control them by cutting off their subscriptions if they are dissatisfied, but in consequence of dispositions made by persons no longer living. When these dispositions become very flagrantly obnoxious to the common weal, or to the sentiment, religious or moral, of the time, they are interfered with by the State in the territory of which they happen to operate, but in spite of this ultimate check, a considerable amount of money is always being spent under the trusts created by deceased persons in ways in which it would not be spent if it was either the property of living individuals or of the State.

The States themselves with their subdivisions exercise a large power of demand. According to democratic theory they should exercise it as directed by the people, but the people are very variously defined by franchise and registration laws, and their power to direct their government is in fact seldom very effective, so that State action is not actually, even in the most democratic countries, merely the joint action of a number of people living on a certain territory.