producer. Now some of the wheat supply only just pays to produce for export, it is said to be grown in "the margin of cultivation"; if it cannot get the former price, it will cease to enter the market. The slightest fall of price has some effect in reducing the supply; there is land in America and elsewhere which even a 2s. duty will put out of wheat-growing, or, what amounts to the same thing, there is land which would have gone into wheat-growing, but is now prevented by a fall of prices. A larger duty of 5s. or 10s. would, of course, reduce the supply of wheat more largely and more rapidly, if it were allowed to remain a tax on the producer. But directly the duty has begun to check the supply of wheat, by making it unprofitable to cultivate some wheat land which was cultivated before, the price of wheat in Great Britain would rise. If there were no "elasticity" in the demand for wheat, ie. if all the wheat supply were absolutely "necessary," the rise of price would not diminish the sale of wheat: all the foreign wheat land must be kept in its former use, and this can only occur, ex hypothesi, on condition that the foreign grower gets his former net price and profit. The attempt, therefore, to throw any of the duty on the foreign producer would fail; the British consumer must pay the entire duty plus cost of collection in higher wheat prices.

The conditions under which the consumer here pays the whole of the tax are seen to be two: first,

the duty is levied on the entire supply of the commodity (there being no home production); secondly, there is no elasticity of demand, a rise of price causing no decline in the quantity purchased.

§ 2. Now precisely because these conditions are not present in the case of a preferential duty upon agricultural imports or a protective duty upon manufactured imports, the duty cannot fall in its entirety upon the home consumer.

Take first the preferential duty upon wheat. As soon as a 2s. duty began to reduce American and other foreign prices for wheat by checking demand, it would begin to check wheat-growing "at the margin," but as it did this wheat prices would begin to rise from, say, the former level of 30s. to 32s. If they reached 32s, the foreign wheat supply for export would remain as before. But the price will not reach 32s. For under the preferential tariff any rise of price over the current 30s. must be supposed to stimulate the cultivation of wheat in Canada, and also to put back into wheat some British land. Though a 25. duty may seem to some to "make no difference," it cannot seriously be doubted that it would turn the balance in some cases, and exercise a proportionate effect in stimulating Canadian cultivation and in preventing some wheat land in Great Britain from going out of cultivation. The notion that a duty, because it is small, may have no effect, is, of course, unthinkable; its effect will only

be slighter and slower than the effect of a 5s. or IOS., into which it will tend to grow.

If, therefore, there is any colonial or British land which can be rendered profitable for wheat-growing or other agricultural use by a rise of price less than is represented by the duty, to that extent the price to consumers will be prevented from rising to the full extent of the duty. The price must rise, for otherwise there is no inducement to the colonial or home grower to produce more wheat than he produces now; it cannot rise to the full extent of the tax, unless we suppose such rise of price to have no effect whatever in stimulating the increase of that part of the supply which enjoys the rise of price without feeling the tax.

The notion, often broached, that the price of our wheat supply is fixed at Chicago by the price set on American export wheat, and that, since Americans would have to recoup themselves for the British duty by a proportionate rise of their prices, the Canadian and even the home producer would reap the benefit of this rise, though in a measure true as regards immediate price changes, has no validity in the long run. Prices of wheat depend in the long run, not on Chicago or other speculators, but upon the conditions of growing wheat on the worst-placed or the poorest land engaged in or available for growing it. If by a preferential tariff we can substitute somewhat worse or less accessible land in Canada or Great

Britain for better land in the United States or Argentina, we shall undoubtedly have to pay a higher price for our wheat, but that price need not, and indeed cannot be, higher by the full extent of the preferential duty.

So the incidence of such a duty would come to be divided between the British consumer and the foreign producer. Some foreign producers would simply cease to produce for our market, others would continue to produce, bearing some part of the duty in the lower prices they got from importing merchants. It is not possible accurately to conjecture what proportion of the duty would fall on producers, what on consumers, without knowing the precise agricultural and industrial resources of all the countries contributing to our supply, so as to be able to calculate the effect of a given rise of wheat prices in driving out of wheat cultivation foreign lands which contribute to our supply, and in stimulating the cultivation of colonial lands. As regards the latter, it is a question, not merely of quantity and quality of land, but of railroad and general industrial enterprise, and of the rate and character of settlement upon new lands. Considering how large a proportion of our wheat and other agricultural supplies we derive from foreign, as compared with colonial and home sources, and the heavy initial outlay in substituting unbroken land in the Canadian North-West for the American fields which now so largely supply

us, it is probable that the price of wheat, at any rate for many years to come, would rise nearly to the limit set by a preferential duty.

§ 3. But the full burden which such a duty would impose upon the British consumer is, of course, not adequately represented in this first result. The prime evil of a protective duty is that it tends to take more from the consumer in enhanced prices than it procures for the treasury. For the higher price is paid upon the part of the supply which does not pay the tax, as well as upon that part which does. If, for example, the result of a 2s. preferential duty on grain were to raise the price Is. 6d. per quarter, while the treasury would only receive £3,300,000, the tax upon the quantity imported in 1902 (amounting to 176,000,000 cwts.), the British consumers would have to pay £6,957,000 in enhanced prices upon the total foreign, colonial, and home supply, amounting to 371,000,000 cwts. Thus the sum paid by the British consumers as a result of the duty would amount to more than twice the sum raised by the treasury through a duty, one quarter of which was paid by the producer, three quarters by the consumer.

§4. In the case of a protective duty of 10 per cent. on manufactured goods the same line of reasoning is generally applicable, though the extent to which the tax can be shifted on to the foreign producer will differ more widely in the several classes of commodities. In the ordinary case, where foreign manu

G