bought for personal consumption means the _satisfaction_ which they
yield, and this is clearly not a thing which is commensurable with
money. When, therefore, it is said that the prices measure their
respective marginal utilities, what exactly is meant? What was it that
the argument of §3 went to show? That the utility of the marginal
pound of sugar would seem to the housewife just worth the price that
she must pay for it; in other words, that it would be roughly equal to
the utility she could obtain by spending the money in other ways. The
respective marginal utilities which _she_ obtains from the different
things she buys will thus be proportionate to their prices. But if she
were to receive a legacy which gave her a much larger income to spend,
she might buy larger quantities of practically every commodity; and,
though she would obtain a greater total utility thereby, the marginal
utility she would obtain in each direction would be smaller, in
accordance with the law of diminishing utility. The prices might not
have changed; the respective marginal utilities to her of the
different things would again be proportionate to their prices, but
they would constitute a smaller satisfaction than before.

Thus we can only say that the prices of commodities will be
proportionate to their real marginal utilities, when we are
considering the different purchases of one and the same
individual. The amounts of money which different people are prepared
to pay for different consumers' goods are no reliable indication of
the real utilities, the amounts of human satisfaction which they
yield. Here we must take account not only of varying needs and
capacities for enjoyment, but of the very unequal manner in which
purchasing power is distributed among the people. The cigars which a
rich man may buy will yield him an immeasurably smaller satisfaction
than that which a poor family could obtain by spending the same amount