THE MARGINAL UTILITY
Thus far we have seen that, utility being the general relation in which
all goods, by their very definition, stand to human wellbeing, value
is that higher, more intimate, more limited relation in which some
particular importance to human wellbeing is conditioned by the having
or losing of some particular good, and a relation of actual dependence
is established between the want and the good. We pass now to the
positive consideration of the measurement of value.
If one good stands over against one want—that is to say, if the
satisfaction of a single want is dependent on the possession of or
power over a single good—there is no difficulty: the value is the
entire utility which the good affords in the given case.
But the estimates of value which practically concern us are not so
simple. We must face the fact that most goods which we have to value
are not single articles, but many goods of the same kind—stocks of
goods—and that, at the same time, most goods are capable of satisfying
several wants. Water, for instance, may be used for drinking, for
washing, for cooling, for ornamental fountains, etc., as books may be
used for reading, for lending, for ornament, for packing, for waste
paper, and so on. But these are uses of very different importance, and
the question is: Which of these utilities is it that determines the
value? This important point cannot be too plainly put, and it will be
wise to follow the Austrian writers generally in taking the risk of
being tedious rather than of being obscure.
A sailor and his dog, the sole survivors from a wreck, have been
tossing on a raft for many days. Land is in sight, but still far away,
and the food is reduced to a couple of biscuits. Both man and dog
are equally famished, and it becomes evident that, unless each gets
a biscuit, one of them will not live to reach the shore. Here we are
confronted with the opposing claims of two wants, that of the sailor
and that of his dog; and, as the sailor is, presumably, the valuer, the
two wants are of very different importance to him. The question is,
What measures the value of the biscuits? According to our formula, the
answer will be found by ascertaining which is the dependent want—which
is the satisfaction that the biscuits condition.
At first sight, one would say that the actual destination of the
biscuits determined this; but that would be to say that two exactly
similar biscuits, both available to the one man, and available under
exactly similar conditions, were of different value. In this dilemma,
one little consideration easily determines the point. If one of the
biscuits were lost, which want would go unsatisfied? For the want which
is satisfied if the good is present, and unsatisfied if it is not, is
evidently the dependent want.[10]
[10] There are two typical cases where valuations are made:—where a
man values something he _has_, with the view of parting with it (in
selling, giving, lending, etc.), and where he values something he _has
not_, with the view of acquiring it. As will be seen from above, the
two methods of valuation come practically to the same result.
The dependent want, in this case, is that of the dog; that is, it is
the _less important_ of the two wants.
To put it now in more general terms. As we saw, the (necessarily)
limited resources at each man’s disposal, he, consciously or
unconsciously, apportions out among his various wants according to his
particular scale, taking care that the more urgent ones are provided
for before the less urgent. It is obvious that, in these circumstances,
there is a _least_ want that is satisfied, although ordinarily we are
not conscious what it is. But it immediately comes to the front when,
from any cause, our resources are diminished. If a working man’s wage
is reduced from twenty shillings to nineteen shillings a week, he
becomes painfully conscious that some want, hitherto satisfied, must go
bare, and the particular want on which he economises immediately points
out which was his least, or least urgent, or final want. In this case,
all the wants previously satisfied are still satisfied except the last
one, and it is proved that none of them depended on having or losing
the shilling. Again, all wants under this, just as before, remain
unsatisfied whether the shilling is there or not. Only this marginal
want is satisfied if the shilling is present and unsatisfied if absent:
it alone, then, is the dependent want.
To recur to our illustration. So long as the sailor had the two
biscuits, one of them would go to satisfying the higher want (his
own), and the other to satisfying the lower want (the dog’s), and
either biscuit was capable of satisfying either want. But, when one
biscuit was lost, the one that remained was instantly elevated to
satisfying the higher want only: it rose, literally, in value because
then it was not a man’s _or_ a dog’s life that depended upon it but
a man’s only: what was lost was the means of satisfying the _dog’s_
want: the less important of the two wants was the dependent one;
and it is the relation of dependence, as we said, that determines
value. We may formulate the proposition thus. The value of a good
is measured by the importance of that concrete want which is least
urgent among the wants satisfied. And we find that what determines
the value of a good is, not its greatest utility, nor its average
utility, nor yet its least conceivable utility, but its marginal
utility in the given circumstances. Jevons called this the Last or
Final Utility. We shall follow Wieser literally in calling it the
Marginal Utility. Simple as this proposition is, experience in teaching
tells me that it is not easily retained so as to be used. For this
reason, it may not be superfluous to confirm its truth by testing it
in various circumstances. I cannot improve on Böhm-Bawerk’s admirable
illustration, and only modify it in non-essential particulars.
A modern Robinson Crusoe has just harvested five sacks of corn. These
must be his principal maintenance till next autumn. He disposes of the
sacks, according to the scale of his wants, in the following way. One
sack he destines for his daily allowance of bread. Another he devotes
to cakes, puddings, and the like. He cannot use more than these in
eating, so he devotes a third to feeding poultry, and a fourth to the
making of a coarse spirit. With these four sacks, we shall say, he is
able to satisfy all the wants that occur to him as capable of being
directly satisfied by corn, and, having no more pressing use for the
fifth sack, he employs it in feeding dogs and cats and other domestic
animals, the companions of his lonely life. The question is: What to
him is the value of a sack of corn? As before, we ask: What utility
will fail him if he lose one sack? It is inconceivable that Crusoe
should have any doubt as to his answer: he will, of course, apportion
out the sacks that remain as before;—two to food, one to poultry, one
to spirits, and he will give up only the feeding of dogs and cats. This
is seen to have been the Marginal Utility—the utility on the margin of
economic employment or use. What he loses, then, by losing one sack
is his former Marginal Utility; and this marginal utility undoubtedly
determines the value of a single one of the five sacks. But here we
come upon another feature of this valuation. If the marginal utility
determines the value of one, it must determine the value of all, as, by
hypothesis, all sacks were alike, and therefore all interchangeable.
Thus we obtain the universal formula for the valuation of goods in
quantity. The value of a quantity of similar goods is the value of the
marginal good multiplied by the number of the goods.
To follow the illustration out. If another sack gets lost, the marginal
utility is found to have been that of the making of spirits; if still
another, the feeding of poultry. Finally, suppose Crusoe to be reduced
to the one sack. Then the satisfying of all lesser wants is out of the
question; the losing of it means death to him; the marginal utility and
the highest utility are one.
Again, suppose Crusoe as merchant bargaining, say, with the Spaniards.
If he have five sacks, he will sell one at a low rate; if he have four,
he will ask a higher price; if he have only one, he will not part
with it for any money. Extend this to the phenomena of an industrial
community. The five sacks represent a larger supply than the four,
the four than the three, and so on; and, as the supply decreases, the
value of the single sack rises. Now one of the commonest phenomena of
a market is that, _ceteris paribus_, increase of supply brings down
value and decrease of supply sends it up. To put it in terms of our
theory: When the quantity of any good produced is increased, the good
is put to lower levels of use; the last want supplied determines the
last satisfaction; and this last satisfaction determines the value
of all the stock. Here we have the explanation of the old Paradox of
Value. If any commodity is available in such quantity that all possible
wants for that commodity are supplied, and yet there is a surplus of
the commodity, the marginal utility is zero, and the value of the
entire stock is nil. And it is also explained how diamonds have a high
value compared with bread. The quantity of diamonds available is never
sufficient to satisfy more than a fraction of the desire for them: the
marginal utility, then, is high. Bread again is, happily, to be had
everywhere at a comparatively small expenditure of labour, and the
immense supply as compared with the limited wants, puts the marginal
utility low.