The Economics of Unemployment/Appendix

The reason why a general trade depression is so dangerous to free trade is that under this condition the ordinary simple refutation of the case for a protective tariff in the immediate interests, not merely of a particular trade but of a particular nation, will not work.

Let us test it by an illustration. At a time of general unemployment in most of the staple industries here and on the Continent, a British railway is considering two tenders for the supply of rails, one from Belgium, one from Glasgow—the former being at a slightly lower price. Now the free trade judgment is absolute in favour of the Belgian rails, as against the protectionist plea for the employment of the idle capital and labour at Glasgow. The free trader argues, not only that it is advantageous to the railway to buy the cheaper foreign goods, but that this course will not reduce the total volume of employment in this country. For, he contends, the purchase of the Belgian rails by English money must, operating through the usual channels of exchange, cause a demand for British export goods which would not otherwise have taken place. There will, he contends, be extra work for otherwise unemployed workers in Northampton, Leicester, Birmingham, etc., to make the goods which go out to pay for the Belgian rails. And it is clear that more employment in the export trades of this country will result from buying the Belgian rails. But does this mean that the total volume of employment in this country will be greater? The protectionist may reply: If the rails are bought from Glasgow instead of Belgium, the same goods which would have gone to Belgium (or their equivalent in other goods) will have to go to Glasgow to pay for the rails made there. The export ​trade will, indeed, be less than it would have been, but the total trade and employment within this country will be greater, because it will contain the making both of the rails and of the other goods which the Glasgow workers will be able to demand as the result of the employment they receive. It is, of course, the familiar argument which Adam Smith adduced under the rather perverse title of "the Two Capitals," urging that there was a prima facie case for preferring internal to foreign trade, because both transactions in the process of exchanging goods against goods were kept inside the same political area. This argument will not, I think, hold water under ordinary trade conditions, in which the available productive power of a country may be considered to be fully employed. Under this condition it is evidently better to buy in the cheaper foreign market, for no total increase of employment will result from preferring the home-made goods. But where there is general unemployment both in this country and abroad, the immediate case for employing otherwise unemployed British capital and labour is far more plausible. "Yes, plausible, perhaps, but not sound," some free traders will certainly reply. A very able English exponent of free trade, the late Mr. Russell Rea, contended that the difference between foreign and internal trade, due to the difference of methods of payment, was such as to invalidate the statement that if the Glasgow rails were bought instead of the Belgian, the same demand for British labour to pay for them would take place. He insisted that the purchase of the Belgian rails stimulated a production of British export goods in payment, for which there would be no equivalent if the purchase of the rails were an act of internal trade. Now I am quite unable to admit that the foreign exchange, operating in one case and not in the other, alters the essential fact that the purchase of the rails enables and impels the recipients of the price to use their purchasing power to demand the production of other British goods. In the one case, the Belgian employers and workers who, otherwise, on our hypothesis of unemployment, would remain idle, will have a job and must spend the money they receive (or cause other Belgians to spend a corresponding sum) in a demand for British goods. In the other case, the Glasgow employers and workers, otherwise idle, will have the job and the money, and will spend it in demand for about the same amount of British goods. The ​fact that the methods of payments are slightly different is irrelevant to the issue. It is curious that it should be necessary to point this out to free traders. For the most fundamental principle of free trade is that a political barrier has no proper significance at all in the sound economy of trade. For the economic international society, to which both British and Belgians belong, it is clearly advantageous that the rails should be bought in the cheaper market, though it means less unemployment in Belgium and more in Britain, but it is foolish to shirk the stark and obvious fact that, if a British railway has the option of setting to work unemployed Glasgow steelworkers on the one hand, or Belgian unemployed steelworkers on the other, the immediate effect of taking the former course is to increase the aggregate employment in this country. The Glasgow workers will buy as much other British produce by the money they receive, and buy it as soon, as would the Belgian workers, and this act of purchase will have the same stimulating effect upon other British trades in both cases. This is quite manifest, if omitting the machinery of payment we confine our attention to the substance of the trade in question. However the payment is effected, what the railway really buys the rails with is the carrying services which it performs. This process of exchanging transport against rails involves no doubt some intermediaries. Probably in neither cases, whether the rails are bought in Glasgow or in Belgium, will the actual railmakers take their payment directly in the transport service of the railway. The railway will sell transport to various British manufacturers and traders, who will pay for it in goods sent to Glasgow or Belgium for consumption by the railworkers in one of these places. However many intermediary processes of sale may take place, the ultimate fact issues that the railway pays for rails in terms of transport, and there is no ground for supposing that the Belgian purchase sets more capital and labour to work at an earlier time than would the Glasgow purchase. The same amount of employment is brought about. The difference is that in the one case part of the employment takes place outside this country, in the other case it does not. Viewing the transaction quite objectively, one may prefer the Belgian purchase, on the ground that, while it makes no appreciable difference in the volume of world employment, it is better for the railway to buy in the cheaper market, and it conduces to the better ​economy of the world that the cheaper or more efficiently conducted business should get the job.

There is, of course, another consideration present to some of your minds which, for simplicity of argument, I have kept back for separate consideration. If the British railway has to pay appreciably more for Glasgow than for Belgian rails, its running costs and freight rates must be raised. This rise of freight might be so considerable as to cripple trade and so strike back at production and employment in the general industry of the country. Such general damage to employment through railway costs might more than outweigh the increased employment given to Glasgow railmakers. On the other hand, if the price difference between the Belgian and the Glasgow tenders were very small, the net result of preferring Glasgow might be a smaller amount of unemployment in this country and a larger in Belgium.

In other words, given a general depression and unemployment in the industrial world, a tariff might be used to distribute the aggregate volume of employment for the time being favourably to the political area which set it up. In the case adduced, a tariff keeping out Belgian rails might, if it did not raise considerably the price of British rails, cause more employment in this country. This judgment is, in reality, only an extension of the common admission that, if a special local trade could keep a special pull upon a tariff, so that its workers gained more as favoured producers than they lost as consumers, their plea for protection had rational validity. On a short perspective, it seems probable that during a general trade depression, a skilfully wielded tariff might shift part of the burden of unemployment on to other political areas. I need hardly add that no mortal tariff would be either skilful or honest enough to keep within the limits of the economy, watching the comparative costs of production in the various countries so closely as to prevent organised industries inside the tariff area from raising prices so high as to cause more unemployment than it cured.

I spoke just now of a short perspective. As a lasting policy such a defence of national employment could not succeed. There would emerge two inevitable causes of failure. The first is primarily political. As in the case of protection for infant industries (quite defensible on theoretic lines) the fatal objection is that such infants never grow up, so here the ​temptation to retain this fiscal remedy for unemployment during normal trade conditions, so as to force disproportionate amounts of capital and labour into certain favoured industries and keep them fully furnished with lucrative orders at the expense of the general body of national trade, would be irresistible. Now the injurious effect of such a policy is obvious. It would offend the first principle of productive economy by forcing into certain avenues of employment capital and labour which could find more productive occupation elsewhere. The total wealth of the community would thus be diminished by spreading the same aggregate of employment in a more wasteful manner. In other words, it is only the recurrent phenomenon of general trade depression which gives specious validity to a tariff as a mitigation of unemployment.

The other fatal flaw in the policy, except on a short range expediency, is economic. It will be admitted, by all free traders at any rate, that the broad effect of any tariff, even under the circumstances here described, must be to force capital and labour to be less productively employed upon the average than if left to the undiluted play of free international trade and investment. This means that such action as we contemplated for the exclusion of Belgian rails would cause rails to be made more expensively in Britain by employing capital and labour in that country than by employing these factors of production in Belgium. If this simple case be expanded into the dimensions of a protective policy, designed primarily to keep the maximum of employment in the protected country, it must fail. For capital employed under these conditions will be less productively and less profitably employed than if it went abroad to Belgium or elsewhere where it was free to utilise cheaper or more efficient labour. The inevitable result of such a protective tariff would be to stimulate the export of new capital from this country, so leaving a reduced supply for the development of our internal industries. The reduced employment for labour thus brought about would stimulate the emigration of workers to countries where the supply of capital and of employment had been made more abundant by the operation of this tariff. The only way of preventing this reduction of her volume of employment in this country would be by preventing or penalising the export of capital, a policy actually proposed by certain persons who ​favour the imposition of a discriminative tax upon incomes derived from foreign investments. But this attempt to keep capital at home, less productively and less profitably employed, would mean a reduction of the average yield of capital, and a consequent, though perhaps not proportionate, diminution of saving for investment. It is important to recognise that an integral and necessary part of a protective policy is this compulsory retention of capital in the protected area, and the consequent reduction of the rate of interest and of the creation of new capital.

Only by putting a sort of ring-fence round a country to stop its capital and labour from flowing abroad in search of more productive uses, as well as to stop foreign goods from entering its frontiers, could protectionism make the appearance of a case for safeguarding employment. Now it is one thing to restrain the export of capital and labour during a brief war emergency, when governmental work ensures full remunerative employment for all of both factors, quite another to erect these restraints into a normal policy, even for dealing with periods of cyclical depression. The conclusion of my argument is this. If it were the sole object of economic statecraft to cause the largest quantity of work to be done within the national area, irrespective of the quantity of wealth produced and the standard of living, a protective policy of this more rigorous order might, if it were confined to periods of general depression, achieve this result. In other words, it is theoretically possible to keep an artificially inflated amount of employment within a favoured area, were it practically feasible or otherwise desirable. A state, in other words, could make its working population ascripti glebæ, keeping them employed at home under a lower standard of production and consumption than they would attain under conditions of laissez faire laissez aller.