Principles of Economics by Alfred Marshall (1890)

Book Three: On Wants and Their Satisfaction

Chapter 5, Choice between Different Uses of the Same Thing. Immediate and Deferred Uses.

1. The primitive housewife finding that she has a limited number of hanks of yarn from the year's shearing, considers all the domestic wants for clothing and tries to distribute the yarn between them in such a way as to contribute as much as possible to the family wellbeing. She will think she has failed if, when it is done, she has reason to regret that she did not apply more to making, say, socks, and less to vests. That would mean that she had miscalculated the points at which to suspend the making of socks and vests respectively; that she had gone too far in the case of vests, and not far enough in that of socks; and that therefore at the points at which she actually did stop, the utility of yarn turned into socks was greater than that of yarn turned into vests. But if, on the other hand, she hit on the right points to stop at, then she made just so many socks and vests that she got an equal amount of good out of the last bundle of yarn that she applied to socks, and the last she applied to vests. This illustrates a general principle, which may be expressed thus: --

If a person has a thing which he can put to several uses, he will distribute it among these uses in such a way that it has the same marginal utility in all. For if it had a greater marginal utility in one use than another, he would gain by taking away some of it from the second use and applying it to the first. (1)

One great disadvantage of a primitive economy, in which there is but little free exchange, is that a person may easily have so much of one thing, say wool, that when he has applied it to every possible use, its marginal utility in each use is low: and at the same time he may have so little of some other thing, say wood, that its marginal utility for him is very high. Meanwhile some of his neighbours may be in great need of wool, and have more wood than they can turn to good account. If each gives up that which has for him the lower utility and receives that which has the higher, each will gain by the exchange. But to make such an adjustment by barter, would be tedious and difficult.

The difficulty of barter is indeed not so very great where there are but a few simple commodities each capable of being adapted by domestic work to several uses; the weaving wife and the spinster daughters adjusting rightly the marginal utilities of the different uses of the wool, while the husband and the sons do the same for the wood.

2. But when commodities have become very numerous and highly specialized, there is an urgent need for the free use of money, or general purchasing power; for that alone can be applied easily in an unlimited variety of purchases. And in a money-economy, good management is shown by so adjusting the margins of suspense on each line of expenditure that the marginal utility of a shilling's worth of goods on each line shall be the same. And this result each one will attain by constantly watching to see whether there is anything on which he is spending so much that he would gain by taking a little away from that line of expenditure and putting it on some other line.

Thus, for instance, the clerk who is in doubt whether to ride to town, or to walk and have some little extra indulgence at his lunch, is weighing against one another the (marginal) utilities of two different modes of spending his money. And when an experienced housekeeper urges on a young couple the importance of keeping accounts carefully., a chief motive of the advice is that they may avoid spending impulsively a great deal of money on furniture and other things; for, though some quantity of these is really needful, yet when bought lavishly they do not give high (marginal) utilities in proportion to their cost. And when the young pair look over their year's budget at the end of the year, and find perhaps that it is necessary to curtail their expenditure somewhere, they compare the (marginal) utilities of different items, weighing the loss of utility that would result from taking away a pound's expenditure here, with that which they would lose by taking it away there: they strive to adjust their parings down so that the aggregate loss of utility may be a minimum, and the aggregate of utility that remains to them may be a maximum. (2)

3. The different uses between which a commodity is distributed need not all be present uses; some may be present and some future. A prudent person will endeavour to distribute his means between all their several uses, present and future, in such a way that they will have in each the same marginal utility. But in estimating the present marginal utility of a distant source of pleasure a twofold allowance must be made; firstly, for its uncertainty (this is an objective property which all well-informed persons would estimate in the same way); and secondly, for the difference in the value to them of a distant as compared with a present pleasure (this is a subjective property which different people would estimate in different ways according to their individual characters, and their circumstances at the time).

If people regarded future benefits as equally desirable with similar benefits at the present time, they would probably endeavour to distribute their pleasures and other satisfactions evenly throughout their lives. They would therefore generally be willing to give up a present pleasure for the sake of an equal pleasure in the future, provided they could be certain of having it. But in fact human nature is so constituted that in estimating the "present value" of a future benefit most people generally make a second deduction from its future value, in the form of what we may call a "discount," that increases with the period for which the benefit is deferred. One will reckon a distant benefit at nearly the same value which it would have for him if it were present; while another who has less power of realizing the future, less patience and self-control, will care comparatively little for any benefit that is not near at hand. And the same person will vary in his mood, being at one time impatient, and greedy for present enjoyment; while at another his mind dwells on the future, and he is willing to postpone all enjoyments that can conveniently be made to wait. Sometimes he is in a mood to care little for anything else: sometimes he is like the children who pick the plums out of their pudding to eat them at once, sometimes like those who put them aside to be eaten last. And, in any case, when calculating the rate at which a future benefit is discounted, we must be careful to make allowance for the pleasures of expectation.

The rates at which different people discount the future affect not only their tendency to save, as the term is ordinarily understood, but also their tendency to buy things which will be a lasting source of pleasure rather than those which give a stronger but more transient enjoyment; to buy a new coat rather than to indulge in a drinking bout, or to choose simple furniture that will wear well, rather than showy furniture that will soon fall to pieces.

It is in regard to these things especially that the pleasure of possession makes itself felt. Many people derive from the mere feeling of ownership a stronger satisfaction than they derive from ordinary pleasures in the narrower sense of the term: for example, the delight in the possession of land will often induce people to pay for it so high a price that it yields them but a very poor return on their investment. There is a delight in ownership for its own sake; and there is a delight in ownership on account of the distinction it yields. Sometimes the latter is stronger than the former, sometimes weaker; and perhaps no one knows himself or other people well enough to be able to draw the line quite certainly between the two.

4. As has already been urged, we cannot compare the quantities of two benefits, which are enjoyed at different times even by the same person. When a person postpones a pleasure-giving event he does not postpone the pleasure; but he gives up a present pleasure and takes in its place another, or an expectation of getting another at a future date: and we cannot tell whether he expects the future pleasure to be greater than the one which he is giving up, unless we know all the circumstances of the case. And therefore, even though we know the rate at which he discounts future pleasurable events, such as spending 1 on immediate gratifications, we yet do not know the rate at which he discounts future pleasures. (3)

We can however get an artificial measure of the rate at which he discounts future benefits by making two assumptions. These are, firstly, that he expects to be about as rich at the future date as he is now; and secondly, that his capacity for deriving benefit from the things which money will buy will on the whole remain unchanged, though it may have increased in some directions and diminished in others. On these assumptions, if he is willing, but only just willing, to spare a pound from his expenditure now with the certainty of having (for the disposal of himself or his heirs) a guinea one year hence, we may fairly say that he discounts future benefits that are perfectly secure (subject only to the conditions of human mortality) at the rate of five per cent per annum. And on these assumptions the rate at which he discounts future (certain) benefits, will be the rate at which he can discount money in the money market. (4)

So far we have considered each pleasure singly; but a great many of the things which people buy are durable, i.e. are not consumed in a single use; a durable good, such as a piano, is the probable source of many pleasures, more or less remote; and its value to a purchaser is the aggregate of the usance, or worth to him of all these pleasures, allowance being made for their uncertainty and for their distance. (5)


1. Our illustration belongs indeed properly to domestic production rather than to domestic consumption. But that was almost inevitable; for there are very few things ready for immediate consumption which are available for many different uses. And the doctrine of the distribution of means between different uses has less important and less interesting applications in the science of demand than in that of supply. See e.g. V, III, sec. 3.

2. The working-class budgets which were mentioned in Ch. IV, sec. 8 may render most important services in helping people to distribute their resources wisely between different uses, so that the marginal utility for each purpose shall be the same. But the vital problems of domestic economy relate as much to wise action as to wise spending. The English and the American housewife make limited means go a less way towards satisfying wants than the French housewife does, not because they do not know how to buy, but because they cannot produce as good finished commodities out of the raw material of inexpensive joints, vegetables etc., as she can. Domestic economy is often spoken of as belonging to the science of consumption: but that is only half true. The greatest faults in domestic economy, among the sober portion of the Anglo-Saxon working classes at all events, are faults of production rather than of consumption.

3. In classifying some pleasures as more urgent than others, it is often forgotten that the postponement of a pleasurable event may alter the circumstances under which it occurs, and therefore alter the character of the pleasure itself. For instance it may be said that a young man discounts at a very high rate the pleasure of the Alpine tours which he hopes to be able to afford himself when he has made his fortune. He would much rather have them now, partly because they would give him much greater pleasure now.

Again, it may happen that the postponement of a pleasurable event involves an unequal distribution in Time of a certain good, and that the Law of Diminution of Marginal Utility acts strongly in the case of this particular good. For instance, it is sometimes said that the pleasures of eating are specially urgent; and it is undoubtedly true that if a man goes dinnerless for six days in the week and eats seven dinners on the seventh, he loses very much; because when postponing six dinners, he does not postpone the pleasures of eating six separate dinners, but substitutes for them the pleasure of one day's excessive eating. Again, when a person puts away eggs for the winter he does not expect that they will be better flavoured then than now; he expects that they will be scarce, and that therefore their utility will be higher than now. This shows the importance of drawing a clear distinction between discounting a future pleasure, and discounting the pleasure derived from the future enjoyment of a certain amount of a commodity. For in the latter case we must make separate allowance for differences between the marginal utilities of the commodity at the two times: but in the former this has been allowed for once in estimating the amount of the pleasure; and it must not be allowed for again.

4. It is important to remember that, except on these assumptions there is no direct connection between the rate of discount on the loan of money, and the rate at which future pleasures are discounted. A man may be so impatient of delay that a certain promise of a pleasure ten years hence will not induce him to give up one close at hand which he regards as a quarter as great. And yet if he should fear that ten years hence money may be so scarce with him (and its marginal utility therefore so high) that half-a-crown then may give him more pleasure or save him more pain than a pound now, he will save something for the future even though he have to hoard it, on the same principle that he might store eggs for the winter. But we are here straying into questions that are more closely connected with Supply than with Demand. We shall have to consider them again from different points of view in connection with the Accumulation of Wealth, and later again in connection with the causes that determine the Rate of Interest.

We may however consider here how to measure numerically the present value of a future pleasure, on the supposition that we know, (i) its amount, (ii) the date at which it will come, if it comes at all, (iii) the chance that it will come, and (iv) the rate at which the person in question discounts future pleasures.

If the probability that a pleasure will be enjoyed is three to one, so that three chances out of four are in its favour, the value of its expectation is three-fourths of what it would be if it were certain: if the probability that it will be enjoyed were only seven to five, so that only seven chances out of twelve are in its favour, the value of its expectation is only seven twelfths of what it would be if the event were certain, and so on. [This is its actuarial value: but further allowance may have to be made for the fact that the true value to anyone of an uncertain gain is generally less than its actuarial value (see the note on p. 135).] If the anticipated pleasure is both uncertain and distant, we have a twofold deduction to make from its full value. We will suppose, for instance, that a person would give 10s. for a gratification if it were present and certain, but that it is due a year hence, and the probability of its happening then is three to one. Suppose also that he discounts the future at the rate of twenty per cent per annum. Then the value to him of the anticipation of it is 3/4 x 80/100 x 10s. i.e. 6s. Compare the Introductory chapter of Jevons, Theory of Practical Economy.

5. Of course this estimate is formed by a rough instinct; and in any attempt to reduce it to numerical accuracy (see Note V in the Mathematical Appendix), we must recollect what has been said, in this and the preceding Section, as to the impossibility of comparing accurately pleasures or other satisfactions that do not occur at the same time; and also as to the assumption of uniformity involved in supposing the discount of future pleasures to obey the exponential law.

Contents | next chapter | Political Economy Archive