But the result might be, that the fall of price might not have an effect
exactly equal, on the demand in the two countries. Suppose, for
instance, that the fall of cloth in Germany owing to the saving of the
cost of carriage, did not increase the demand for cloth in Germany; but
that the fall of linen in England from a like cause, did increase the
demand for linen in England. The linen imported would be more than could
be paid for by the cloth exported: the difference must be paid in money:
the change in the distribution of the precious metals between the two
countries would lower the price of cloth in England, (and consequently
in Germany), while it would raise the price of linen in Germany, (and
consequently in England). Germany, therefore, by the annihilation of
cost of carriage, would save in price more than the cost of carriage of
the cloth; England would save less in price than the cost of carriage of
the linen. But if by the miraculous annihilation of cost of carriage,
England would not _save_ the whole of the carriage of her imports, it
follows that England did not previously _pay_ the whole of that cost of
carriage.

Thus, the division of the cost of trade, and the division of the
advantage of trade, are governed by precisely the same principles; and
the only general proposition which can be affirmed respecting the cost
is, that it is _pro tanto_ a deduction from the advantage. It cannot
even be maintained that the cost is shared in the same proportion as the
advantage is; because the increase of the demand for a commodity as its
price falls, is not governed by any fixed law. Suppose, for instance,
that the advantage happened to be divided equally: this must be because
the greater cheapness arising from the establishment of the trade,
either did not affect the demand at all, or affected it in an equal
proportion on both sides. Now, because such is the effect of the degree
of increased cheapness resulting from importation burthened with cost of