employment will be much less than that which has been supplied with work, so that a net increase of British employment will result.

The protective policy, in this analysis, appears to have transferred a certain amount of employment from foreign capital and labour to British capital and labour, not, indeed, appreciably altering the total amount of unemployment in the industrial world, but reducing the proportion of this waste which falls on British industry.

It is the theoretic possibility of filling this hole of unemployment by an artificial diversion of trade through protection that gives what plausibility attaches to a tariff. It might be a profitable economic policy for a nation to pay a slightly higher price for certain consumable commodities, and by reducing to a corresponding extent its demand for other commodities to displace capital and labour in these industries, if the result were to secure full regular employment for considerable quantities of capital and labour which would otherwise be wasted.

But between this theoretic service of protection and a practically serviceable tariff there is a great gulf fixed. For it must be borne in mind that the theoretic validity of this remedy for unemployment rests wholly on the hypothesis that there exists a considerable margin of spare available power of capital and labour, an unemployed margin beyond the legitimate needs of ordinary business adjustments,

and that, therefore, orders diverted from the foreigner could be executed without causing capital and labour to leave employments where they were more productively engaged.

But it is impossible to know how large this margin is in any given trade at any time, and any tariff which stimulated a British industry beyond the limit of this margin would cause grave injury, by diverting capital and labour already productively employed to a less productive employment. Since the logic of protection constantly drives in the direction of greater stringency and wider extension, it would be impossible to restrict a tariff to a certain carefully measured protection for certain trades; trades already protected would demand more protection, other trades at first unprotected would clamour for protection, and no effective stand could be made against either sort of pressure.

Talk of the application of a "scientific" tariff is illusory; the ever-changing conditions of commerce and of the industrial arts render delicate "scientific" forecasts impossible, and when the interests of a trade are strongly engaged in producing an impression, and in supporting that impression by apparently substantial evidence, they will always get the better of the "scientific" expert. Their art will conquer his science. Their strong self-interest will overpower his weaker guardianship of the further

interest.

This is inevitable. The two instances

where we have admitted the theoretic validity of protective measures are open to this perversion in a peculiar degree. The first is the case of a tax on foreign goods dumped on our shores with the object of damaging our home industry, seizing our market, and then raising prices. This tax we found might be economically efficacious and politically advisable, if it could be correctly imposed. But we found that the proof of evil intention, which is the condition of the application of this remedy, would be well-nigh impossible; the British industry seeking protection could hardly ever bring direct evidence of intention on the part of foreigners, and would be virtually confined to providing evidence of mere underselling, due possibly to better economy of production or of discrimination between home and export prices, not in itself an injury to the recipients of this bounty.

The second instance of theoretically valid protection is the case we have here under discussion. When we turn from the general and theoretic to the individual and concrete, we find ourselves confronted with similarly insuperable difficulties. A British trade competing with foreign imports is losing ground; it cannot get sufficient orders to keep its capital and labour fully occupied. British orders are going to foreign firms; a good many mills are closed or working half time; many workmen are on the unemployed list. Here is a case where it seems feasible

by a tariff on imports to secure an increase of the aggregate of employment for Great Britain in accordance with our analysis. But this assistance is only valid on two conditions: first, that the additional work thus secured can be executed by the unemployed margin without trenching on capital and labour usefully engaged elsewhere; secondly, that the industry thus helped is not a decaying industry. This last point requires further elucidation. Some British industries will at any given time be growing rapidly and absorbing more capital and labour; others will be virtually stationary, their capital and labour sometimes fully occupied, sometimes not, according to the general condition of trade; a few industries may be actually declining, either because of a definite superiority of some foreign industry, or from change in fashion affecting demand, or because they are displaced from their position in the export cargo of Great Britain by other commodities. If any of these causes is operative it signifies a natural decline of a British industry necessary and desirable in the interests of British industry as a whole; the trade is not worth maintaining in its former shape and size. Now such a genuinely decaying industry will necessarily show a large quantity of unemployed capital and labour. The social utility has in reality gone out of this; the forms of capital should be left to die, and the labour should be assisted to find other employment as quickly as possible. To support such

an industry by means of a tariff would be a public injury, impeding the wholesome processes of internal readjustment in the industry of the nation. But if a tariff were once admitted as a remedy for unemployment, it would certainly be abused for the support of these decaying industries. It would be practically impossible in most instances for officials to determine between true decline and the conditions which we have designated as an excessive margin of unemployment.

§7. When to these genuine difficulties of discrimination we add the political abuses rampant under tariff legislation, the case against the use of a scientific tariff as a remedy for "unemployment" becomes overwhelming. Nowhere in tariff legislation or administration can "science" or "theory" hold its own against the political "pull" of industrial interests. In the United States the "scientific theory" is that the tariff should be so arranged as to protect American capital and labour against the competition of cheaper European production by equalising the expenses of production in Europe and America.

But it is not seriously pretended that the tariff is actually designed to conform to this idea. The plasticity of a tariff law in process of construction is such that it reflects much less the economic needs than the political power of the various in

M