Some Aspects of the Tariff Question

Taussig, Frank William
Display paragraphs in this book containing:
First Pub. Date
Cambridge, MA: Harvard University Press
Pub. Date
1st edition.

Part I

Part I, Chapter I

Duties, Imports, Prices


In this introductory chapter I shall consider, even at the risk of repeating elementary matter, the way in which duties work, the significance of the continuation of imports after duties have been imposed, and the possibility of measuring the charge which they lay on the community.


A common notion is that any duty operates automatically as a price-raising cause, bringing at once and permanently a tax up to its full rate. Not a little speechifying of a very effective kind has consisted in an enumeration of extreme rates, with the implication that they bring burdens no less extreme.*1 Even more remarkable is the eagerness with which protected producers have themselves schemed and labored for high duties as if it were certain that they would get the full benefit, in a corresponding rise in the price of their wares. The burden which our protective system imposed on the community has been much exaggerated by its opponents; but the protected producers and their spokesmen have countenanced the exaggeration by virtually endorsing the indictment against themselves. They ask for advances in duties and protest against reductions as if a corresponding effect on domestic prices were certain to appear.


The truth is that the levy of a duty may have no influence at all on domestic price; or it may raise the price of the dutiable commodity by its full amount; or it may have an effect intermediate between these extremes.*2


(1) The first case is the simplest. A duty on a commodity which is produced within the country as cheaply as without, and sold as cheaply, ordinarily has no effect whatever. Of such levies there has been a plenty in our tariff history. Those on the staple agricultural products are the most familiar and conspicuous. In the log-rolling which is an almost universal concomitant of protective tariffs, the notion that a duty will surely be of benefit to domestic producers has caused our farming sections to insist on "their share" of the going favors, and to accept, nay demand, duties on wheat, corn, meat and meat products, which yet have been quite without industrial effect. There has been no more striking illustration of the average farmer's naïve state of mind on this subject than the bitter opposition aroused by the reciprocity treaty with Canada which the Taft administration proposed in 1910-11. The free admission of wheat contemplated by that treaty was supposed to portend disaster to the wheat growers of the northwest; though it was known to all the world that wheat was exported both from the United States and from Canada, and that it was the same in price (allowing for cost of transportation) in these two countries and in England. The range of commodities subjected to duties yet not at all affected by them, has been very wide, including not only agricultural staples, but many manufactured articles.


(2) The second case—that in which the price of the commodity rises by the full amount of the duty—is found when imports continue after its imposition. Nevertheless it is not so easy as may seem at first sight, to determine just how conclusive is the evidence from the fact of importation. It will appear, as we proceed in the discussion, that qualifications of various sorts need to be borne in mind.


Under the ordinary conditions of trade,—those of competitive dealings,—the continuation of imports after a duty has been levied shows that the price of the commodity is higher within the country than without by the full amount of the duty. This is not the same as to say that the price is raised to the purchaser or consumer by that full amount; a consequence which no doubt commonly ensues, but by no means ensues under all conditions. It is conceivable that the divergence between foreign and domestic price will come about through a fall in the foreign price, not through a rise in the domestic; or through a partial fall in the one, a partial rise in the other. Of this possibility, more will be said presently. The only thing which is shown by the inflow of imports over a tariff barrier is that the level of price is higher on one side than on the other by the height of the barrier. The reason is obvious: no trader will import goods, and pay the duty on them, unless he can sell them at an advance over the foreign price which will recoup him for the duty paid.


This holds, to repeat, under the ordinary conditions of trade. But it does not hold necessarily in case of goods produced in the foreign country under monopoly conditions. Under monopoly, there is a possibility of difference in charge to different purchasers, and hence a possibility that a duty will not affect price as it would under the conditions of a free market.


That the incidence of a tax on monopoly products is different from that of a tax on competitive products is a commonplace in economics. Whether the tax be in the form of an excise or a tariff duty, the monopolist may find it expedient to bear part of the charge, in an extreme case even to bear the whole of it. He may be confronted by such inelasticity of demand as to make it most profitable to sell at an advance in price less than the tax, perhaps very much less than the tax. Now, the peculiarity of a customs duty is that it makes divided markets. It is imposed not on the whole of the monopolist's output, but only on that part which is exported to the country levying the duty. In the duty-levying country, the monopolist may not raise his price by the full amount of the duty,—i.e., may lower his net price, what is left to him after the tax is paid by himself or others,—and yet maintain the full previous charge in his home market. Then the divergence between the two markets after the duty has been imposed will be less than the amount of the duty. The foreign producer then will "pay" some part of the tax; not in the sense that he lowers his price all around, but that he lowers it on the quota exported to the duty-levying country.


An analogous case is that of "dumping" in its typical form,—that is, the steady sale of a commodity to a foreigner at a lower price than to domestic consumers. The divergence of prices is here also explicable, as a rule, on the ground of monopoly. Of dumping in its various forms, more is said elsewhere;*3 it is enough at this stage to note that it presents theoretical problems very similar to those of the imports of a monopolized article continuing after the imposition of a duty.


Complete monopoly is rare at best; and this particular consequence of full monopoly seems to be even more rare. I know of no case, either in American or in foreign experience, where one having a complete monopoly has in fact continued steadily to send to a foreign country a product on which a duty has been levied, and then has sold the product at an advance in price less than the duty.


But imperfect monopolies,—those where the product is sold for a considerable time at a price above the strictly competitive rate,—are by no means rare. Competition works out its effects slowly and irregularly. For long periods there are quasi-monopolies due to established reputation, trademark, or brand. No doubt these require for their maintenance, as well as for their first establishment, a considerable degree of business ability; but they are susceptible of being held in a position of advantage for a surprisingly long time. As in the case of complete monopoly, though to a less extent, the returns are so high as to make it possible to make some reduction in price and yet retain enough to make sales worth while. The imposition of a duty may lead to such a concession. Thus, a particular kind of steel tape used by engineers, made in England and widely exported, has a long-established name and a quasi-monopoly position. "Specialties" of various kinds are brought from Europe to the United States under similar conditions, and indeed account for the continued importation of many classes of goods subject to high duties. On such articles the reduction in price which may follow the imposition of a duty is not likely to be great; the divergence between foreign and domestic price will after all not be far from the amount of the duty. But the cases are frequent enough, and divergence sufficiently noticeable, to cause the man of affairs who encounters them to be skeptical about the general proposition that price rises by the full extent of the duty even when imports continue; and they lead the protectionist to jeer once more at the "theoretical" free trader who says that the foreign producer bears no part of the tariff burden.


The case is different with commodities produced under strictly competitive conditions. Here there is a free market, and a market price the same for all purchasers. Here it would seem that there is no possibility of divergence between prices to different purchasers, such as appears in case of monopoly. It would seem, therefore, that the continuance of imports proves at the least that price in the duty-laying country is higher than in the exporting country by the full amount of the duty. Here, too, it would seem clear that in the long run this full amount constitutes a charge on the domestic consumer, not on the foreign producer. These consequences do in fact appear; yet with temporary divergences which again puzzle the ardent free trader and are made much of by the ardent protectionist.


A manufacturer or set of manufacturers whose operations have been developed and adjusted for a large export trade may be "caught" by the sudden levy of a duty. In order to hold their own in a market on which they have relied for disposing of a large output, they may sell in the duty-laying country at a less price than elsewhere. One would suppose that, under competitive conditions, the concession in price would not be confined to the exported quota. Each of the producers,—so the economist would reason,—is desirous of avoiding the fall in price; each will prefer to sell in the home market at full price rather than in the foreign market (now subjected to duty) at a reduced price. Competition between them will cause the decline to be distributed over all the output. But in fact things often work out, at least for some time, differently from what the close-reasoning economist expects. The producers, it is true, are desirous of staving off the fall in price; but this desire often leads them, without any express agreement or combination among themselves, to maintain their price on ordinary sales, yet cut it perforce on the sales to the protecting country. They do not wish to "spoil" the general market*4 or upset the going price which has come to be regarded as "fair." Thus for a time the consequence may be similar to that of monopoly; there may be a reduction from the going price, for the purchasers in the duty-laying country. But all this is for a time only. Such special sales at reduced prices are unwelcome; they will be dropped as soon as possible. Each producer will prefer to sell all he can in the general market where concessions in price have been avoided. In this general market, too, he will be tempted to push his sales; very probably by concessions other than overt reduction of price,—such as longer credit, ready allowance for alleged damage or shortage, assumption of freight charges. The mercantile world has plenty of devices by which rates are cut in fact, even though nominally maintained. The differences between the prices in sales to the duty-laying country and to other markets will gradually disappear; and then, if imports into the former go on, the normal inference from continued imports can be drawn: price is higher within such a country than without by the full amount of the duty.


Further, that difference will ultimately appear as a charge on the domestic consumers, not on the foreign producers. Only for a time will the latter sell in the duty-laying country at a less (net) price than they have been previously getting,—assuming that this previous price was the strict competitive price. Sooner or later they will withdraw from the business thus made unprofitable or less profitable. If this cannot be done without an appreciable reduction in total output, the process will require time; most of all, if it cannot be done without allowing large plant to wear out.*5 But in the end special sales to foreign countries and general reductions in price due to the cutting off of the foreign markets, will cease. Exports which may be sent thereafter to the duty-laying country will go under normal conditions, and the normal consequences of duties will appear. Prices of the dutiable commodities will be higher by the full amount of the duty; not only higher within the protecting country than without, but higher by that full amount when measured from the previous level. In the long run, the continuance of imports of staple goods, after a duty has been imposed, proves that the domestic consumer pays an enhanced price, or tax, to the full extent of the duty.


In the present volume, it happens, the discussion of imports, duties, prices will have to do chiefly with staple goods made under competitive conditions; moreover, goods not mainly dependent on the American market, so that even a temporary divergence from normal conditions will rarely need to be considered.*6 As regards the tariff schedules to be considered in the following pages, the general proposition holds, with little need of allowance for the qualifications and exceptions: if imports continue, we may be sure that the domestic purchaser pays a tax of the full amount of the duty.


If now imported goods, steadily sent in over a tariff barrier, are raised in price by the amount of the duty, it follows that any similar goods that may be produced within the country are also raised in price by the same amount. Not only the imported supply, but the total supply, sells at a price higher by so much within the country than without. This is the first article, and an essential one, in the free traders' indictment of protective duties: they tax the consumer without bringing a corresponding revenue to the government. They thus cause prima facie a net loss to the community. The higher price paid for the imported portion is not open to this charge; what the consumer so pays in taxes is offset by the revenue yielded to the public treasury. It is the higher price of the domestic product which has no offset. All this is a commonplace in economics, and there is no occasion for repeating here what has been so often set forth.


(3) Next we have to consider the third case, intermediate between the two just discussed,—that in which a duty causes a rise in price, but one not up to its full amount. Here the duty is prohibitory, yet has its effects. It is so high as to cause the cessation of imports which would otherwise come in. The case is one in which there is "need" of protection; the commodity could be got more cheaply from abroad; but the duty is greater than is needed to offset the difference between domestic and foreign cost. There is then no overt evidence on the quantitative effect of the duty. The tax on the domestic consumer may be nearly equal to the full amount of the duty; it may be considerably less. So far as the evidence from imports goes, there is nothing to prove there is any tax at all,—the case might be that mentioned first in our analysis.


The intermediate case is the most frequent of all as regards manufactured goods. It is not often that a duty is imposed on these precisely so high as to cause a division of the market between foreign and domestic producers. Such a result was aimed at in our tariff act of 1913, in which the rates were supposed to be adjusted on a "competitive" basis.*7 In fact, a rate that is really "competitive" is difficult to fix, and was arrived at in very few of the duties of 1913. A duty on a manufactured product commonly is either so high as to keep out all imports, or so low as to admit all and thus to be in effect merely a revenue duty. True, imports often appear to continue, and a division of the supply between domestic and foreign quotas often appears to be brought about. But the appearance is deceptive; the two sets of goods on examination prove to differ in quality, or to be for other reasons not in reality competitive. Of the need of discrimination in interpreting the evidence from continued imports of manufactured goods, more will be said in the ensuing paragraphs, and still more in the later chapters of the volume.


A duty on the so-called raw materials is more likely to be really competitive; the probability is greater that some part of the supply of such goods will be brought in over the barrier of a duty. The reason for this difference between manufactured goods and extractive products is not far to seek. The latter are likely to be produced not at uniform cost, but higher cost for some parts of the domestic output than for others. When a duty has brought about a rise in domestic price, there will be some increase of domestic production, but not an indefinite increase. Diminishing returns, i.e., increasing cost, will set in, and will bring a limit to the extension of the domestic quota. Imports will continue, even though on a less scale than they would without a duty. Of this situation there have been some striking illustrations in modern tariff history. One was in the continuing imports of wool into the United States during the period from the close of the civil war until wool was admitted free in 1913; a case which will be followed in detail in this volume.*8 Another was in our imports of raw sugar, of which also a full consideration will follow. Still another, the occasion of a vehement political and economic controversy in Germany, is in the sustained imports of wheat into that country after the imposition of the wheat duty in 1879 and its gradual increase in the years thereafter.*9 In all these cases the fact that imports came in steadily after the imposition of the duty proved beyond question that the price of the whole supply, domestic as well as foreign, was raised by the full amount of the levy.


But, to repeat, in the case of manufactured goods, of which an increased supply can be produced in the long run without rising cost per unit, the division of the market between foreign and domestic producers is not so likely to take place. It may be fairly described as a lucky hit when a duty is adjusted at the exact point which brings about this result. In the tariff experience of the United States at large, and particularly as regards the schedules whose effects will be examined in this volume, the rates have usually been much above the point of prohibition. Imports have ceased. To ascertain then what effect the duties have had, above all to measure their quantitative effects, proves extremely difficult. Statistics of the prices of the goods are not easy to get, and are even less easy to compare with due allowance for differences in quality. In some instances, as with ordinary grades of cotton cloths, it is tolerably certain that domestic prices have been no higher than foreign; the case is in reality our first. With the ordinary grades of woolens, on the other hand, it is clear that domestic prices have been higher than foreign, yet by an amount much less than the duty; the case is the intermediate one. And for another great class of textiles, silk fabrics, the evidence is conflicting and the outcome difficult to state with any precision; there is a conglomerate made up of the two extreme cases, and of various degrees of the intermediate case.


Returning now to a topic touched in passing a moment ago, we have to note some further cautions and qualifications to be observed when drawing inferences from the fact of continued imports. There are not a few cases where imports seem to prove the full rise in price, but in fact do not prove anything of the kind.


In the first place, it must be ascertained whether the goods imported are in reality comparable to those made within the country. Textiles of all sorts have been steadily imported into the United States during the period covered in the present volume,—cottons, woolens, silks. But the imports have been almost exclusively of the finer and more expensive qualities. The less expensive goods, those which are most largely used, have been made exclusively within the country. The consumers have been served by two streams of heterogeneous supplies, not by one of homogeneous supply. Though the custom house statistics register considerable imports of silks and woolens, these have been of grades and qualities different from the domestic goods.


A striking case is that of pig-iron. Of this article also the customs returns show imports in considerable quantities for each year during the last half-century. But for the greater part of the period they were of special qualities only; classed as "pig-iron" in the tariff schedules and in the Treasury statistics, yet in fact without significance in the general iron market. Almost all of the imports were of spiegel-eisen and ferro-manganese, used in comparatively small amounts for mixing with other iron in the Bessemer process. This continued importation proved something about the relation between foreign and domestic price for that particular grade, but nothing about the prices of the enormously greater quantity of pig-iron proper.*10


Again, exceptional transportation conditions may cause an imported commodity to find its way into some part of the domestic market over a duty which yet is prohibitory as regards the general market. Steel rails may be carried from Great Britain to Galveston, by steamers which are glad to get a return freight for cotton, at very low transportation charges; and it may then be to the purchaser's advantage to import them and pay the duty (i.e., a price raised by the amount of the duty) rather than meet the comparatively high land freights from the American mills,—at Pittsburgh or Birmingham (Ala.). Yet steel rails may be as cheap in Pittsburgh as in Great Britain, and American prices for them in general not higher than British. So economical is water transportation that steel rails have been transported from Europe around Cape Horn to Puget Sound, and have paid a considerable duty, even though rails were in most parts of the United States no dearer than in Europe. Similarly, pig-iron might come from Glasgow to New England and other places on the Atlantic coast, though charged with a duty and though no higher in price at Pittsburgh than at Glasgow. Transportation conditions of this kind explain some continuing imports which have puzzled those who make inferences from the bare statistics of foreign trade.


Lastly, we have to consider another qualification and distinction. It is one thing to say that the continuance of imports proves domestic price to be higher than foreign price by the full amount of the duty; it is another thing to say that the domestic consumer pays a tax to that full amount. The latter proposition, usually stated without qualification by the free traders, is often denied by protectionists of the extreme type. These are likely to maintain that duties operate as taxes on the foreign producer, not on the domestic consumer. To say that duties always tax the foreign producer is absurd. Yet there are conditions,—quite apart from monopoly, or temporary conditions of readjustment,—under which the unqualified free trade statement is not completely true, and the extreme protectionist statement not completely false; conditions under which imports continue, price is higher by the full extent of the duty, yet the domestic consumer is not taxed to that full extent. And conversely there are conditions under which a remission of duty will not lower price by the full amount.


These are the conditions, familiar in economic theory, where production is carried on under varying cost or diminishing returns. The ordinary free trade reasoning, like most of the reasoning of those British economists by whom the theory of international trade was worked out, assumed constant returns,—one uniform cost of production, irrespective of the volume of output. This at least was assumed as regards the foreign supply. The influence of varying cost or diminishing returns on domestic supply, and the consequent special effects of import duties on domestic cost and on the rent of land, were conspicuous in the reasoning of those who attacked the British corn laws. But these same conditions may exist for the imported supply. Suppose that the imports are of agricultural products or raw materials, and that they come from a country whose natural resources are not superabundant. An increase in the output of a commodity so produced will cause its normal price to go up, if the additional increments of supply can be got only at higher cost. A decrease in output, conversely, will cause normal price to go down, if the sources of supply which are abandoned are comparatively poor, and if those which continue to be utilized are comparatively good. The margin of cultivation will rise in the former case, will fall in the latter; and normal price will shape itself correspondingly. The particular case which is to be considered in the present discussion is where an import duty causes part of the foreign supply to be supplanted by domestic supply, and where the abandoned foreign quota had been produced at high cost. The recession of the margin of cultivation will then cause normal price to fall in the foreign country; and though imports continue, and though domestic price be higher by the amount of the duty, it will not be raised by that full amount above the level which prevailed before the duty was imposed. It cannot be said that in this case the foreigner bears any part of the tax; but, also, it cannot be said that the domestic consumer pays a tax of the full amount of the duty. The converse case arises where a duty which had long been imposed and had shut out a foreign supply, is repealed, letting in the foreign article. If the consequent pressure on foreign sources of supply causes resort to poorer grades of land or other natural agents,—if the margin of cultivation goes up,—the normal foreign prices will rise. Then, in the country where the duty has been remitted, price will go down by less than the amount of the duty. Some part of the possible gain to consumers will be offset by the higher cost of the additional foreign supplies.


This sort of general reasoning, however, is in fact less likely to be applicable to imported supply than to domestic supply. The British economists who made much of it in condemning the corn laws, but neglected to consider its applicability to the countries from which corn might be imported, were substantially in the right, even though their theoretic reasoning was not carried far enough. It is much more probable that the conditions of diminishing returns will be found for a domestic supply than for a foreign supply. The reason is obvious. The available area in any one country is more likely to be limited, and, therefore, more likely to exhibit considerable variations in cost. A foreign supply is likely to come, actually or potentially, from several countries. Within wide limits, it will probably be produced under conditions not of varying cost but of constant cost. Any considerable increase in the supply of wheat grown in Germany or in England, for example, will probably cause resort to inferior soils, or disadvantageous pressure on all the available soils. But the same increase of supply from foreign countries,—distributed over Canada, Argentina, the United States, India, Russia, Roumania,—will cause no pressure at all. If indeed a single country or area were the sole source of supply for the article subjected to duty, there would be some probability of increasing cost and rising price after the removal of the duty. But this must be a rare case; at all events I know of none in the tariff experience of the United States. More nearly within the bounds of possibility is the case where, though several countries contribute to the imports, all of them have pushed production to the point where additional output is not certainly to be had on the same terms. This possibility exists, for example, in the case of wool; and it has been alleged to exist, though with less plausibility, in that of sugar. As will appear later, it deserves at least to be considered whether a greater demand for foreign wool, due to the abolition of the United States duty, will cause some permanent rise in foreign cost and price, and so fail to bring for the domestic consumer the full expected gain from the remission. Even in this case the answer seems to be in the negative: the conditions of foreign supply are sufficiently flexible to prevent an outcome so disappointing to the free traders.*11


So much for the details, qualifications, exceptions, which must be borne in mind when interpreting statistics of imports or reasoning about the effect of duties on domestic price. Under the ordinary conditions of trade, if imports continue, the effect of a duty on prices is plain. The nature of the effect is equally plain, though its extent is not so easy to measure with exactness, if imports are stopped by the duty; yet would come in were the duty removed. Quite a different question is whether these consequences from the imposition of a duty are permanent; whether the price of the dutiable article, raised at first by the tariff, may not be lowered eventually in consequence of changes in the conditions of domestic production. This is the question raised by the doctrine of protection to young industries, to which we turn in the next chapter.*12

Notes for this chapter

One of the familiar methods of enumeration is to describe the taxes which follow the consumer from the cradle to the grave; a modern use of this tactical device is in the speech of Mr. Underwood, when introducing the tariff bill of 1912-13 in the House of Representatives, August 13, 1912.
In this analysis I follow the method of Albert Gallatin, in his Free Trade Memorial of 1831; reprinted in the collection which I have edited, State Papers and Speeches on the Tariff, pp. 122-123.
See chapter xiii, pp. 202-212.
Cf. the remarks on a similar situation in Marshall's Principles of Economics, Book V, chapter v, § 6 (6th ed.).
A case of a different kind, yet analogous to those considered above in that it rests on abnormal conditions, is adduced by Professor Brentano. The Russian state, being under obligation to make heavy remittances to foreign countries on interest account, reduced its railway rates on rye when Germany raised duties on that grain; thus virtually shouldering the duty. The stolidity, lack of resource, and general immobility of the Russian peasantry are also said by Brentano to have contributed, for a considerable time, to the same result. L. Brentano, Die deutschen Getreidezölle, p. 22.
An exception to this statement appears in the case of tin plate. There the foreign (British) production had been largely for the American market, and the duty of 1890 did serve for some to depress the British price. See below, chapter xii, p. 176.
See my Tariff History of the United States, p. 418 (edition of 1914).
See chapter xix.
On this episode full figures are given in Brentano, Die deutschen Getreidezölle.
Cf. what is said below, chapter x, pp. 144 seq.
See below, chapter xix, p. 318.
The reader will note that I speak in this chapter merely of the difference between price with the duty and price without the duty, not committing myself on the question whether this difference constitutes or measures a national loss. The presumption is that a national loss occurs, and is measured by the enhanced price which the consumer pays on the goods produced at home (not on those imported, since here the consumer's burden is offset by the government's revenue). Those conversant with the theory of international trade need not be told that there is the further possibility that duties will disturb the equilibrium of international demand and supply, and lead to a readjustment by which the duty-levying country will gain. See the classic passage in Mill, Political Economy, Book V, chapter iv, § 6. Cf. Marshall's Memorandum (of 1908) on the Fiscal Police of International Trade, §§ 7-9; Taussig, Principles of Economics, chapter xxxvii, § 1. The possibility has been questioned, but not in my opinion on solid grounds, in a note by H. H. O'Farrell in Quarterly Journal of Economics, August, 1912. Some further theoretical aspects of the problem seem to me to deserve attention; but this is not the place for examining them.

End of Notes

Return to top