F. W. TAUSSIG

INTERNATIONAL TRADE

1927

————————————

Chapter 1. Three Cases
Chapter 2. Absolute Differences in Cost
Chapter 3. Equal Differences in Cost
Chapter 4. Comparative Costs
Chapter 5. Wages and Prices in Different Countries. Domestic Prices and International Prices
Chapter 6. Wages Not Uniform—Non-Competing Groups
Chapter 7. Capital and Interest
Note on the method of handling capital and interest
Chapter 8. Varying costs; diminishing returns; increasing returns
Chapter 9. Varying Advantages
Chapter 10. Two Countries Competing in a Third
Chapter 11. Non-Mehchandise Transactions Tributes, Indemnities, Tourist Expenses
Chapter 12. Non-merchandise Transactions Further Considered Loans and Interest Payments, Freight Charges
Chapter 13. Duties on Imports and the Barter Terms of Trade

————————————
————————
————

[3] AT the outset I must ask the reader's patience, and a withholding of judgment on his part, till the close of a systematic and prolonged exposition. Some assumptions will be made in the earlier chapters that must appear quite out of accord with facts, and some extremely simple and apparently unreal situations will be considered. It will be assumed, for example, that countries or regions trade with each other by a process of barter, and that the people concerned weigh deliberately the advantages of barter. It will be assumed, again, that within each country commodities are exchanged, by those making them, on the basis of the quantities of labor given to their production. These commodities, too, will be assumed to be produced at constant cost—that is, tho the volume of output may change, each unit is always produced by the same quantity of labor. And so on thru a long list of provisional suppositions. As the subject is unfolded step by step, these will be qualified and supplemented, and verisimilitude will be attained, or at least some approach to verisimilitude. Many comments and criticisms that must arise in the reader's mind will be met, I trust, as he proceeds. He is asked to judge the analysis which follows as a whole, regardless of seeming inadequacy in any one part considered by itself.

We begin by distinguishing three sorts of cases. To these may be applied the terms: (1) absolute differences in cost; (2) equal differences in cost; (3) comparative differences in cost.

In the analysis and interpretation of these cases, "cost" will mean labor cost, measured in terms of time,—so many days. [4] A given number of days' labor, supposed necessary for producing a given number of commodities, will constitute the cost of the commodities. It is not money cost or expenses that we start with, but labor and effort. Cost is low when little labor is needed to produce a given physical quantity, high when much is needed. It is inverse to the effectiveness of labor: the greater that effectiveness, the lower is cost.

I am aware that the point of departure thus chosen is open to criticism. It is that of Ricardo, Mill, and their successors; and the treatment of the subject at the hands of this school has always had a certain air of unreality, as if divorced from the actual conduct of trade. Instead of beginning with labor cost, as they did, we might first consider cost as understood in the everyday world—money cost, or supply price—and proceed thereupon to the labor cost. Either course is justifiable. The choice between them is one of methodology, or rather of the better method of exposition. In the chapters that follow the relation between the two kinds of cost, and their respective bearings on international trade, will receive careful consideration. The realities will not be neglected, even tho the first approach may seem far removed from the channels of actual trade. And the conclusions eventually reached would have been the same, and would have been reached by essentially the same reasoning if the starting point had been money cost, and if there had then been a working back to labor cost. The question, as just remarked, is one of procedure. I will not pretend that the plan of the present book is clearly the better. The reader is asked at this stage merely to bear it in mind.

For the present, then, we define cost in terms of quantity of labor, not in terms of money cost or of supply prices. On this basis we differentiate the three cases. Representative figures, of the kind which will be freely used in the ensuing pages, are as follows:

Case I. Absolute Differences

[5] In the United States the effectiveness of labor in producing copper is twice as great as is that of labor in producing copper in Germany; cost is one-half as great. In Germany, on the other hand, the effectiveness of labor in producing linen is twice as great as it is in the United States for that article; cost (of linen) again is one-half as great. Each country has an absolute advantage over the other in the production of one of the commodities.

Case II. Equal Differences

Here cost is lower in the United States than in Germany—the effectiveness of labor is greater—as regards both commodities, and in equal degree for both. The ten days produce 30 of copper in the United States, 20 in Germany. The ten days produce 15 of linen in the United States, 10 of linen in Germany. Labor is more effective in the United States all around by fifty per cent.

Case III. Comparative Differences
(a)

(b)

Case III has two variants. In variant (a) the American effectiveness of labor is greater than the German in copper (30 to 20); whereas in linen there is no difference (15 in both countries). Considering the whole situation, the United States is the more fortunate country; but it is her better fortune in copper that alone makes the difference.

In variant (b), however, the United States has lower cost and greater effectiveness for both commodities, but not to the same [6] degree for both. For copper the ten days in the United States produce 30, in Germany 15; the difference is as 2 to 1. For linen the ten days produce in the United States 15, in Germany 10; the difference is as 3 to 2. The United States has a greater advantage in copper than in linen. She may be said to have a comparative advantage in a more special sense.

In the chapters that follow we proceed to consider what are the possibilities of trade between the two countries in these three cases, and what the possible terms of trade. The situation will be analyzed first as if the conditions were the very simplest. Suppose the trade to be one of barter, the direct exchange of copper for linen. The two countries may be supposed to get together in mass meeting, so to speak, and to consider whether anything can be gained by an exchange of goods; much as we should consider what might be done by two collectivist communities which had no common medium of exchange.

[7] WE begin with Case I. Consider first what would be the terms of trade within each country if both of the commodities were produced in each and were exchanged within each. Assume, that is, that the people of the two countries have nothing to do with each other. They go their own ways, quite without contact or trade between them. Assume further that the trade within each—this purely domestic trade—takes place as it would if the labor in each moved with complete freedom from occupation to occupation, from industry to industry. The two commodities then are exchanged, in the United States and also in Germany, in proportion to the amounts of labor needed to produce them. In other words, ' we proceed, as regards exchanges taking place within a country, on the basis of a labor theory of value. It will appear in due time how important is this assumption, and in what ways it needs to be revised and qualified. For the present, we simplify the case by supposing not only barter between the two countries, but also barter within each on terms settled by equalization of the returns to labor.

The figures for Case I, to repeat, are:

therefore in the United States 30 copper = 15 linen.

therefore in Germany 15 copper = 30 linen, or 7-1/2 copper =15 linen,

There is a wide diversity between the terms of trade that would obtain within the two countries. In the United States 15 of linen would exchange for 30 of copper; in Germany 15 of linen for 7-1/2 of copper.

[8] If now our supposed mass meetings took place and the possibility of trade between the two countries were contemplated, it is obvious that Germany would gain by sending 15 linen to the United States and getting anything more than 7-1/2 of copper in exchange. The 15 of linen are in Germany the product of 5 days of labor (30 linen for 10 days, hence 15 linen for 5 days). That same amount of labor in Germany (5 days) would produce 7-1/2 of copper. By shipping 15 of linen to the United States and getting more than 7-1/2 of copper in exchange, Germany gains. Conversely, the United States would gain by an exchange of 15 linen for anything less than 30 of copper. The ten days labor produce in the United States 30 of copper; any less quantity of copper (25 or 20) is the product of a correspondingly smaller amount of labor. But 10 days in the United States are needed to produce 15 linen. If the United States gets the 15 of German linen for as much as 27, 28, 29 of copper, she still gets the linen for less labor than would be needed to produce it at home.

The quantitative relation between these physical amounts— between pounds of copper and yards of linen—we designate as the "barter terms of trade." We are assuming, be it remembered, that no money is used and that the transactions are simply and solely the exchange of goods for goods. Much copper for little linen, or much linen for little copper—these are the possibilities of the barter terms of trade. What is the meaning of the phrase under this simple condition is obvious enough and is easily visualized.

But with the use of money and with sales and purchases in terms of money—under the complex conditions of the actual foreign commerce of the world—the facts which the phrase describes are by no means easy to visualize. Not only are they very difficult to keep in mind, but they are so disguised, so overlain by other more conspicuous and not less significant phenomena, that one is apt to forget that they exist at all. In the every-day talk about foreign trade their existence is completely ignored; one would not dream that there was such a thing. Even in the discussions of the economists they are often forgotten. Yet this is the important thing; it is here that we have the essence of international trade;

[9] it is here that the fundamental problems appear. Let the reader attend to the phrase "barter terms of trade," prepare for its recurrence, bear in mind its meaning.

A word of explanation on this matter of phraseology may be helpful. "Terms of trade" was suggested by Marshall.⁰⁰¹ "Terms of exchange, which at first may seem better, is confusing, for the reason that "exchange" is so constantly used in connection with the price of "bills of exchange" and the rates of "foreign exchange." Unfortunately there is nothing in the English tongue that corresponds to the German "Devisen." As in so many other problems of economics, we have retained the ways of common speech, and we designate by the same word—"exchange"—quite different things. The possibilities of misunderstanding will be lessened if we use the phrase "terms of trade" to designate the fundamental process of the traffic of goods for goods, and restrict the use of "exchange" as much as possible to bills of exchange and to dealings in them. To reduce still further the possibilities of misunderstanding, I have modified Marshall's phrase in the manner stated above, and shall speak of the "barter terms of trade."

Returning to our analysis, we simplify it by neglecting cost of transportation. The most obvious effect of cost of transportation is to narrow somewhat the range within which the terms of exchange are confined. If it costs one of copper to carry each batch of copper from the United States to Germany, and also costs one of copper to carry each batch of linen from Germany to the United States, the limits within which the exchange can take place will be 8-1/2 and 29 of copper for 15 of linen, not 7-1/2 and 30. At any figure above 8-1/2 Germany will still gain; at any figure below 29 the United States will still gain. There are some further and less obvious complications from the introduction of cost of transportation, to which the older economists gave much attention and of which something will be said as we proceed.⁰⁰² All the complications from this factor, obvious or obscure, do not affect the essentials of [10] the analysis, and for the present will be ignored. We shall assume that cost of transportation is so small an item that it may be neglected. Each commodity is supposed to exchange for the other on the same terms in both countries; which would not be the case if transportation costs were taken into consideration.

Each commodity then will be produced in one of the countries only. Copper will be produced in the United States only, being sent thence to Germany in exchange for linen. Linen will be produced in Germany only, being sent thence to the United States in exchange for copper. No linen will be produced in the United States, no copper in Germany.

If now we suppose the two regions are not separate countries whose inhabitants are kept apart by differences of language, custom, race, but are parts of a single homogeneous country—say New York and Pennsylvania, not Germany and the United States—our conclusions become different. They become different, that is, so far as concerns the possible barter terms of trade. The terms will then be settled at the fixed rate of 15 linen for 15 copper. The inhabitants of both regions will participate equally in the potential gains from the trade. But as regards trade between countries—to return to this case—under a rate more favorable to the United States, say of 15 linen for 8 copper, the total rewards of producers in that country, in quantities of linen and copper together, will be greater than in Germany. On the other hand, under a rate more favorable to Germany, say 15 linen for 29 copper, the total rewards will be greater in that country. If there be free movement of people between the two, such as takes place between different regions within a country, there will ensue equality of reward. That equality is reached under the rate of 15 linen for 15 copper.

The circumstance that gives to international trade its outstanding characteristics is the lack of free movement between the trading countries, and the consequent possibility of great divergencies between the returns to labor in them. The further significance of this factor, the causes of divergences between countries, the nature and effects of the divergences which still persist within [11] a country, will be considered more at length as we proceed. The first and the main consequence in international trade is indicated in the simplest form by the figures just given. Barter terms of trade may emerge, and rewards to labor under which much more advantage accrues to one country than to another. And, to repeat, under the conditions here supposed, the difference in gain to the countries may be very great. The United States may gain a great deal from the trade and Germany but little; or Germany may gain a great deal and the United States but little. The barter terms may be highly favorable to Germany and but little favorable to the United States; or may be highly favorable to the United States and but little favorable to Germany.

We proceed now to modify the suppositions still further for Case I, in such way as to come a step nearer the realities. Countries do not exchange products for products thru mass meeting votes or any other process of conscious bargaining. Goods are sold for money. To quote Ricardo's phrase, "every transaction in commerce is an independent transaction." Linen will be sent from Germany to the United States only if it sells for a less price in Germany; and copper will be sent from the United States to Germany only if it in turn sells for a less price in the United States. Our next step is to introduce the mechanism of money and prices.

In doing so we still simplify the case. Suppose the circulating medium in the two countries to be the same, and to be gold. Assume that copper and linen are both sold for gold in the two countries, and that gold as well as the commodities moves freely from one country to the other.

Go further with the process of preliminary simplification. Assume not only that gold is the currency of both countries, but also that the greater or less plenty of gold money affects prices. More money means higher prices, less money lower prices. We shall assume, in other words, the validity of what is called the quantity theory of money. It is not material whether we accept also another proposition which goes with the quantity theory, namely that prices rise or fall in precise proportion to the increase or decrease of the monetary supply. It suffices for the present [12] reasoning that they do rise when money becomes more plentiful, do fall when money becomes less plentiful. An inflow of gold from one country to another causes prices to fall in the first, to rise in the second. I would not be supposed to imply that so bare and simple a statement as this tells the whole story of the working of the monetary mechanism. As will appear later, some of the most troublesome complications of the subject arise precisely in connection with the working of that mechanism. The reader's patience is asked once more. Let the simpler aspects of the problem be first cleared up.

Suppose now money wages to be $1.50 a day in the United States and $1.00 a day in Germany. And suppose too that this is the sole expense involved in producing the commodities in the two countries. Ignore any return to capital; to that complication, as to others, attention will be given in due course. With regard to the supposed figures on wages, consider them for the present merely as a status qua. Accept them as existent. It will appear presently how they came to exist, and in what way such money rates of wages are related, in the last analysis, to the prices of the goods.

Then, still assuming that within each country there is free movement of labor and that goods sell on the basis of the labor involved in producing them, we have the following results:

Observe the term "domestic supply price." By this is meant the price at which, under the given conditions of expenses of production, each article could be steadily brought to market and sold. It is the (simplified) money cost of production; the kind of "cost" which figures in most economic discussion and in all business discussion. Since we are here (and in almost all parts of this book) using "cost" in the other sense—that of labor cost—it will obviate misunderstanding and will conduce to clarity if [13] we do not use the word at all when referring to the forces that act on money prices. In considering these we shall therefore speak of domestic supply prices only; or, for brevity, supply prices. The German domestic supply price of copper is that at which copper would sell in Germany if there were no international trade at all, if copper were produced in Germany, and if its price were purely a domestic matter. It is the price which the German producers must get in order to induce them to make that article. Similarly the domestic supply price of linen in the United States is the price which would rule in the United States if there were no trade between the countries—the price which Americans must get in order to induce them to make linen at all.

A cursory inspection of the figures shows that the supply price of copper is lower in the United States ($0.50 there, $0.66-2/3 in Germany) while that of linen is lower in Germany ($0.33? against $1.00 in the United States). The American producers of linen would be undersold by the German producers of linen; the German producers of copper would in turn be undersold by the American producers of copper. American copper would be sold in Germany, German linen in the United States. The United States would produce no linen, Germany would produce no copper.

Suppose now the difference in money wages between the two countries to be the other way—higher in Germany. Suppose German wages to be $1.50, American wages $1.00. Then we have:

Here the same movement of goods takes place. Copper is cheaper in the United States than it could be in Germany ($0.33-1/3 there as against $1.00 in Germany) and will move thence to Germany. Linen is similarly cheaper in Germany ($0.50 against $0.66-2/3) and will move thence to the United States.

In other words, it is quite possible that money wages will be higher in the United States or that they will be higher in Germany. [14] And evidently the gain from the exchange will be greater to that country which has the higher money wages. The price of each commodity, once the exchange has begun, will be the same in both countries. The German price for linen will be the ruling price in the United States as well as in Germany. The American price for copper will be the same in Germany as in the United States. But, with money wages higher in the United States and with the prices of the goods identical (both linen and copper), evidently the American purchasers will be better off. Conversely, if money wages be higher in Germany, as they quite possibly may be, the Germans will be better off.

The range of possible deviation in money wages between the two countries will be between $1.00 and $2.00. It is possible that money wages in the United States will be nearly double those in Germany; and money wages in Germany may be nearly double those in the United States. If, for example, money wages are $2.00 in the United States and $1.00 in Germany, we have:

Here the supply price of copper is the same in the two countries—$0.66-2/3. But linen is very much cheaper in Germany—$0.33-1/3 against $1.33 in the United States. No copper will move from the United States to Germany; but linen will move from Germany to the United States. As German linen is sold in the United States, gold will have to be remitted in order to pay for it. The outflow of gold from the United States will lower prices there, and will lower money wages also; while the inflow of gold into Germany will raise prices and money wages. Thereupon copper will begin to move in payment for linen. How great the movement will be and what the limits to it, need not at this stage be considered. It is enough to observe that as soon as money wages in the United States become less than $2.00—less than double the German rate—the possibilities of sales of goods both ways [15] will arise. And the converse applies, of course, to Germany. If German wages are twice as high as in the United States—$2.00 there against $1.00 here—the price of linen will be as low in the United States as in Germany, and copper alone will move between the two countries. With German wages less than $2.00—less than twice as high—linen will move from Germany, and an exchange of goods for goods will begin.

In analyzing the possibilities of exchange under the supposition of barter, we have seen that the terms on which the barter of goods for goods will take place may vary between wide limits. It is easily shown that the same possibility of wide variation in these terms exists under a money regime also.

Let us suppose money wages to be higher in the United States than in Germany, but not as much higher as they might possibly be. Let wages in the United States be $1.50, wages in Germany $1.00: not twice as high, but 50 per cent higher. Then we have:

Linen is produced more cheaply (in money) in Germany, and flows thence to the United States; copper more cheaply in the United States, and flows thence to Germany; linen obviously being relatively the cheaper of the two.

The barter terms of trade are 15 of copper for 22-1/2 of linen. At the price of copper which rules in both countries ($0.50) the sum of $7.50 will buy 15 of copper; while at the ruling price of $0.33-1/3 for linen, that same sum will buy 22-1/2 of linen. In other words, 15 of copper sent from the United States to Germany will procure 22? of linen. And these barter terms of trade are, of course, in accord with the money wages in the two countries. Money wages in the United States are 150 per cent of those in Germany, and the product of ten days of American labor (30 of copper) exchanges for the product of 15 days of German labor (45 of linen).

[16] The converse supposition—equally permissible under the given conditions—is that money wages are 50 per cent higher in Germany. Let them be $1.50 in Germany, $1.00 in the United States. Then we have:

As before, linen is produced more cheaply in Germany, and flows thence to the United States; copper more cheaply in the United States, and flows to Germany; but now with copper, not linen, relatively the cheaper of the two. The barter terms of trade are now 15 of copper for 10 of linen, or 22-1/2 of copper for 15 of linen. At the price of copper which rules in both countries ($0.33-1/3) the sum of $7.50 will buy 22-1/2 of copper; while at the ruling price of linen ($0.50) that same sum will buy 15 of linen. The barter terms of trade have become more advantageous to Germany. And this betterment of Germany's gain is in accord with her higher money wages, now 50 per cent above those of the United States.

Suppose now that money wages are the same in the two countries. Let them be $1.50 alike in Germany and in the United States. Then we have:

Copper moves from the United States to Germany, and linen from Germany to the United States. Exchange of goods takes place, both countries gain, and both countries gain alike. The barter terms of trade are no more favorable to the one country than the other, they exchange their labor par for par, so to speak. Money wages being the same, the inhabitants of each country have the same degree of benefit from the trade between them.

[17] This last supposition is in accord with what we should expect to find, and in the main do find, in trade within the bounds of a homogeneous country—in domestic trade. If we suppose the two regions to be, for example, New York and Pennsylvania, equality of money wages may be expected, and thereby equality of gain from the trade between them. Marked divergence from such equality would lead to a movement of population from one to the other, and eventually to an equal sharing of the gain. This result is dependent, it need hardly be said, on effective homogeneity between the peoples of the two regions; the inhabitants of both are able and willing to move as freely between the regions as within them, and can apply their labor as effectively in the one as in the other.

It is hardly necessary to point out that such an equal sharing of gain, with equality in money wages, does not imply an equal distribution of population between the two regions. New York or Pennsylvania might have very different numbers. That one of them which produced the commodities most in demand would have the larger population; the other would have the smaller. The volume of output both for copper (say) in Pennsylvania and for linen in New York would adjust itself to the conditions of demand for these articles in the two states, and the distribution of workers between them would adjust itself accordingly. The free movement between them which brought about the equality of wages would bring about (probably) an unequal distribution of population and a distribution shifting with changes in demand. It is the absence of free movement, and the consequent fixity of each population, that causes something other than equality of remuneration to determine the share which each region shall secure of the possible gain from trade. That other factor, as will appear in the sequel, is the state of demand in the two countries for their several products.

The type of international trade considered in this chapter—that in which there are absolute differences in cost—is found mainly in the trade between tropical and temperate countries. The tropical countries have an absolute advantage, because of climatic conditions, [18] in their several products; the countries of the temperate zone, and especially those of European culture, have similarly an absolute advantage, tho resting perhaps not so preponderantly on physical causes. That this is the sort of trade in which a gain accrues unmistakably to both sides has long been admitted. So obvious, indeed, is the gain that no attempt has ever been made, at least by the peoples of the temperate regions, to hamper it by protective duties or other restrictions. But tho this much is readily perceived, few understand that trade of this kind offers not merely a possibility of clear gain, but a possibility of greatly varying apportionment of the gain; that it might be carried on under conditions quite different from those familiar to us, and still be to mutual advantage; and that the key to the apportionment of advantage is found in the money incomes of the people of the exchanging countries. The most striking concrete illustration is in the trade between Great Britain and British India. The trade is free (some recent restrictions thru import duties are so slight as to be negligible). Money incomes are high in Great Britain, low in India. Those goods which are exchanged between them—international goods—sell at virtually the same prices in both. Evidently the Englishman, with his high money income, is in a better position as purchaser of these international goods than is the East Indian with his low money income. But this is no necessary feature of the trade. It is quite conceivable, and quite consistent with the continuance of the trade, that the situations should be reversed: that the East Indian, not the Englishman, should have the higher money income and the greater share of the possible gain. And it is no less conceivable that money incomes in the two regions should be the same, and the gain thus shared equally. The existing situation is not at all a necessary outcome from the given conditions, but only one among several possibilities; and the significant indication of the nature of the outcome which in fact has been reached is the difference of money incomes between the exchanging countries.

[19] THE figures representing Case II, it will be remembered, are:

Cost in terms of labor is lower in the United States for both articles; labor is more effective all around by 50 per cent. Then we have:

It is obvious that no gain can accrue to either country from a direct exchange of goods for goods. In both, linen and copper, if produced within their respective limits, would exchange on the same terms. The labor that would produce 15 linen in the United States would produce in that country 30 of copper. Precisely the same is the case in Germany. If Germany were to take the product of 10 days' labor in linen (10 of linen) and carry this amount to the United States, she would there get, at the American terms of trade, 20 of copper; and that amount of copper the same 10 days of labor will enable Germany to produce. Similarly, the United States could gain nothing by sending copper to Germany for exchange against linen.

It is further obvious that the people of Germany would presumably find it to their advantage to transfer themselves en masse to the United States. I say presumably; the advantage would not necessarily be realized. True, Germany is the less prosperous country; labor is less effective all around than in the United States; [20] and it would seem that her people have but to move to the better region in order to enjoy its better conditions. So it would be if the cause of advantage in the United States were merely of a physical sort—climate, land, or what not—and if the Germans were quite as capable of utilizing the natural conditions as the Americans. But if the differences in effectiveness do not rest purely on physical grounds; if they are due to aptitudes which the Americans possess but the Germans do not; if the Germans on moving to the United States could not there apply their labor with the same intelligence, ingenuity, vigor, as the Americans,—there would be no certainty of gain from the shift. As we shall see, causes of this kind do operate. Not physical causes alone—natural resource and the like—determine the current of international trade; the human factor counts heavily.

At this stage of our inquiry, however, we assume that the Germans remain in their own country. It is not material for what reason they do so. It may be that they are indifferent to a possible gain from transplanting themselves, being attached to their own country, familiar with its mode of life and ignorant of life abroad, immobile even tho they might prosper by moving. They are to be supposed, for the purposes of the present argument, to go their own way regardless of the American possibilities; the Americans also go their own way. And then the two groups will have no occasion for exchange of goods.

The same conclusion is readily reached under the supposition not of barter, but of money and prices: sales of goods and transactions between individuals.

We may make at the start any supposition whatever with regard to prices and wages; it will remain to be seen what figures represent the definitive conditions. Suppose that in the United States wages are $2.00 a day, in Germany $1.00 a day. Then we have:

[21] The prices of both goods are lower in Germany than in the United States. Copper and linen alike would be exported from Germany. Specie would flow from the United States to Germany; prices would fall in the United States and would rise in Germany.

Suppose now that wages are still $2.00 in the United States, but are at the same rate in Germany—$2.00. Then we have prices in the United States as before. But now —

All prices now are higher in Germany than in the United States. Both copper and linen would be sent from the United States to Germany, specie would flow from Germany to the United States, prices would fall in Germany and rise in the United States.

Lastly, suppose wages in the United States to be $2.00, in Germany $1.33-1/3. Then we have (for ready comparison the United States figures are presented again):

The price of each commodity is the same in the two countries. Copper sells for $0.6-2/3 in the United States and in Germany too. Linen sells for $1.33-1/3 in Germany and in the United States too. Neither article will move from one country to the other. There is no trade between them.

In other words, under the conditions of equal differences in cost, money wages will adjust themselves exactly to these differences. The effectiveness of labor in the United States is to that in Germany as 3 to 2; money wages in the United States are to those in Germany as 3 to 2. Prices of goods are the same in the two countries. The people of the United States, with their higher money wages, will be able to buy more goods for the wages of a given amount of labor—one day's or ten days'—than the Germans can buy with [22] their wages for the same amount of labor; this being the concrete form in which the greater material prosperity of the United States shows itself. There is no gain to anyone from moving goods between the two countries. Labor might possibly move to advantage but goods cannot move. And so each country goes its own way, each with its own scale of money incomes, both with the same scale of prices.

[23] WE turn now to Case III, that of differences in comparative costs. For convenience in following the numerical illustrations, I shall use in the exposition that follows a different set of figures from those of the preceding chapters. The same principles are involved; the figures now used are selected for their greater simplicity.

Let us suppose that

Here the United States has an advantage in the production of both commodities. The effectiveness of her labor is greater both in wheat and in linen, but it is not greater to the same degree in both. In wheat it is as 20:10, while in linen it is as 20:15. The American effectiveness of labor in wheat is twice as great as that of Germany; in linen it is one-third greater. There is in the United States a comparative advantage in the production of wheat.

We might denote the situation by a different set of phrases. Instead of speaking of a comparative advantage, we might say that the United States has a superior advantage in wheat. She has an advantage in both commodities, but one greater in wheat than in linen. Germany, on the other hand, has an inferior disadvantage in linen. She is at a disadvantage in producing both commodities; but the disadvantage is less in the case of linen than in that of wheat. The outcome of an inferior disadvantage in trade between countries (or for that matter between individuals, or between groups within a country) is precisely the same as that of a superior advantage. [24] They are two forms of comparative advantage, each the complement of the other.

Looking again at the figures of our case, it is obvious that in the United States 10 wheat would exchange for 10 linen if both articles were produced within the United States. In Germany 10 wheat would exchange for 15 linen if both were produced in that country. At any rates between, both countries would gain from an exchange. At an exchange of 10 wheat for 14 linen the United States would gain 4 of linen, since the 5 days' labor entailed by producing the 10 of wheat would produce in the United States not 14 of linen, but only 10. At the rate of 14 linen for 10 wheat, Germany too would gain, but not so much; her gain would be 1 of linen. If on the other hand the terms of trade were 10 wheat for 11 linen, the United States would gain 1 linen and Germany 4 linen. At any rate of 10 wheat for more than 10 linen and less than 15 linen, both would gain. And at any such intermediate rate, each would confine itself to that commodity in which it had a comparative advantage. The United States would produce only wheat, and would get all her linen by sending wheat to Germany, exchanging it for linen; she would confine herself to the industry in which she had the superior advantage. Germany would produce only linen, and would get all her wheat by sending linen to the United States and getting wheat in exchange; she would confine herself to the industry in which she had the inferior disadvantage.

Turn now from this supposition to that of a money regime. Suppose that wages in the United States are $1.50 a day; wages in Germany are $1.00 a day. These figures, be it noted again, are merely illustrative; they do not stand for a necessary or definitive outcome. Assuming them for purposes of illustration, we have further:

[25] Observe that the prices of the goods are such that wheat moves from the United States to Germany, linen from Germany to the United States. Wheat, tho produced in the United States by labor receiving wages higher than those in Germany, yet sells for $0.75 in the United States. Wheat, if produced in Germany would sell there for $1.00; its domestic supply price—its cost of production in the business sense—is higher in Germany even tho it is produced by labor receiving wages lower than those in the United States. This sort of situation is apt to surprise most people: how can an article sell at a low price in a country where wages are high? The answer is simple enough: it can do so if the high paid labor is effective, as in this case with American labor for wheat. On the other hand, Germany can produce linen at a lower price than the United States, notwithstanding the fact that German labor is less effective in producing linen than American. The explanation again is simple: the labor, tho less effective, is paid at a money rate which is not only less, but is lower to an extent more than in proportion to the less effectiveness. Wheat is cheaper in the United States, tho produced with high-paid labor; linen is cheaper in Germany, tho produced with ineffective labor.

Come now a step still closer to reality. What quantities of goods might be expected to move between the two countries, and what total sums of money might they represent in the way of imports and exports ?

Suppose—again for illustration —

The money sums balance; the wheat sent from the United States exactly suffices to pay for the linen received from Germany. The transactions will be carried out thru the mechanism of the foreign exchanges, and the demands for bills of exchange would be exactly met by the offerings of exchange. Exchange would be at par, no specie would flow between the two countries; all is quiet.

With these transactions, 8,000,000 bushels wheat exchange for 9,000,000 yards linen. The barter terms of trade thus are 8 wheat = 9 linen; or, 10 wheat = 11-1/4 linen. That the trade takes place [26] on these terms is a matter of which no one in either country is conscious, unless indeed it be some sophisticated economist. All that appears on the surface is that some Americans sell wheat in Germany and some Germans sell linen in the United States; and that the two amounts are equal in the only terms of which most persons think—in money. That money wages are higher in the United States is also a circumstance of which they are aware; but they take this as a matter of course, as a sort of God-given relation, not to be further inquired into. The fundamental fact, which quite escapes their attention, is that in terms of physical quantity the people of the United States get 11-1/4 of linen for every 10 of wheat which they send to Germany.

Observe what will be the conditions of prosperity—the incomes in terms of commodities—in the two countries. Assume that in both one-half of the money wages is expended on wheat, one-half on linen. Then,

The American is able to buy with his day's wages, in addition to the one bushel of American wheat, 1-1/8 yards of German linen; just 1/8 more than he could buy if the linen were made in the United States. The German is able to buy, in addition to 3/4 yard of German linen, 2/3 bushel of American wheat, whereas he would have been able to buy only 1/2 bushel if the wheat had been of German production. The American gains 1/8 yard of linen from the international exchange; the German gains 1/6 bushel of wheat.

It is conceivable, however, as was indicated by the previous analysis of barter conditions, that the people of the United States might exchange with Germany on better terms. They might get—so it has been indicated—more than 11 linen for their 10 of wheat, up to a maximum of nearly 15 linen. Under what circumstances will they get more?

The Germans, under the above conditions of equilibrium, have been assumed to take 8,000,000 bushels wheat from the United States [27] at the price of $0.75. Suppose, however, that at this price they want more than 8,000,000; the German demand for wheat increases. More than the 8,000,000 bushels are bought in Germany, and more than $6,000,000 become payable by German purchasers to Americans. Foreign exchange no longer is at par. Exchange rises in Germany, falls in the United States. Specie flows from Germany to the United States. Prices tend to rise in the United States, to fall in Germany; and not only prices, but money wages also.

The nature of the outcome of these readjustments is indicated by such figures as the following:

At these rates of money wages:

Observe that, as before, the prices of both articles are such that they will move from country to country. Wheat in the United States at $0.85 still sells for a lower price than that at which it can be produced in Germany ($0.90). Linen at $0.60 in Germany still sells for a lower price than that at which it can be produced in the United States ($0.85). The range of difference between the domestic supply price and the import price is now less in wheat than in linen; but the differences are of the same kind as before.

The physical quantities moving between the countries also will be affected. More wheat will move from the United States to Germany, obedient to the greater German demand. But more linen will also move from Germany to the United States. Money incomes as well as prices rise in the United States, and the Americans will be tempted to buy more German linen. Since prices (as well as money incomes) fall in Germany, the price of linen will [28] fall and Americans will be tempted still more to buy German linen. On the other hand, as American wheat prices rise, the German purchasers, tho still buying more wheat than before, will gradually slacken in their purchases of the American article. The final outcome will be something like the following:

The money sums balance. Imports exactly pay for exports in each country. Foreign exchange is again at par; no specie flows.

The barter terms of trade have now become more favorable to the United States. In the physical quantities of the goods a different relation has developed 10-1/2 million bushels of wheat now exchange for 15 million yards of linen; the terms of trade are 10-1/2 for 15, or 10 for (nearly) 14-1/4. Under the previous supposition the United States got 11-1/4 linen for every 10 bushels of wheat; now she gets 14-1/4 linen for every 10 bushels.

Continuing the analysis, we may examine what would be commodity wages ("real" wages) in the two countries. If, as before, the purchasers in both spent one-half of their incomes on each commodity, we should have:

American wages have risen in terms of linen. Before, the day's commodity wages were 1 wheat plus 1-1/3 linen; now they are 1 wheat plus 1-5/12 linen. German wages will have fallen in terms of wheat; the day's commodity wages, which had been 3/4 linen plus 2/3 wheat, now are f linen plus 9/17 wheat. The less advantageous terms on which the trade takes place for Germany are reflected in lower commodity incomes for the Germans; and the more advantageous terms for the United States are reflected in higher commodity incomes for the Americans.

[29] These last figures, however, call for correction. They have been worked out on the supposition that, as before, one-half of the wages in each country are spent on linen, one-half on wheat. But this supposition is not consistent with what has been assumed with regard to the conditions of demand; namely, that the demand of the Germans for wheat has increased. The Germans, under that assumption no longer spend one half of their wages on wheat. They have come to spend a larger proportion of their total income on wheat. The Americans, on the other hand, have been tempted by rising prices of wheat to send more of that commodity to Germany, and have been led to consume less of it themselves. And they have also been tempted by falling prices of linen to buy from the Germans more of that article; they have come to spend a larger proportion of their income on linen. We need to modify our figures accordingly.

Make the modification by supposing that under the changed conditions of demand 3/5 of the Americans' income is spent on linen, 2/5 of it on wheat; and that in Germany 1/3 of the income is spent on linen, 2/3 on wheat. Then we have:

Compare now the figures with those arrived a ton page 26, where the commodity wages were worked out under the first set of conditions, before the increase in German demand for wheat. Because of the increase in demand the nature of the changes in commodity wages is as follows:

Can it be held, under this revised (and consistent) statement, that commodity wages in the United States have unquestionably risen, those in Germany unquestionably fallen ? True, the Americans [30] get for the day's labor more linen than before—1-7/10 linen instead of 1-1/8. But they also get less wheat—8/10 instead of 1. The gain in linen is considerable, the loss in wheat comparatively small. But are wheat and linen commensurable? And is it certain that the Americans secure a net gain from having more of linen when they also have less of wheat? The Germans, on the other hand, while they have less of linen—only 5/10 linen instead of 3/4—have a bit more of wheat: they have 12/17 instead of 2/3. For them, also, is it certain that the loss in linen over-balances the gain in wheat ? Their loss in linen is considerable, the gain in wheat is slight; but the question for them, also, is how can these changes in the quantities of different commodities be reduced to common terms?

The only answer is that in terms of "utilities" (satisfactions, enjoyments, gratifications) both Germans and Americans have gained. In ultimate income, in terms of the psychological outcome, all are better off than before. This is shown by the simple fact of choice. As is assumed in the entire hedonistic calculus, it is choice between alternatives which alone shows whether we "like" or "enjoy" one source of satisfaction more than another. The Americans get less of wheat than they got before; but they have been led to accept that less amount because they have been tempted by the cheapness of linen. The fact that they give up some wheat in order to get more linen proves that the rearrangement suits them better, that their income in terms of satisfactions is greater than before. The Germans, too, while they have less linen than before, have more of wheat. Had it not been for their desire to have more wheat, they would not have parted with so much of their own product in order to get so small an addition of the American product. Given their changed demand, they are better off than before; they must like the present situation better than the earlier one or it would never have developed.⁰⁰⁴

The sense in which it can be said, then, that the new conditions [31] of exchange are unfavorable to the Germans is simply that in physical units more of the German product is given for the American than before. In this sense, and in this only, are the barter terms of exchange less favorable to Germany than before.

Returning now to the main trend of the argument, we may state as follows the general conclusions which are indicated. The barter terms of trade between the countries depend on their states of demand; on the intensity and elasticity of Germany's demand for wheat, of America's demand for linen. The limits within which the terms of trade are confined depend on the range of difference in costs. In the case supposed, 10 of wheat can exchange for any amount of linen between 10 and 15. But the mere analysis of the differences in costs indicates nothing about the precise point between the limiting extremes at which the exchange will take place. That depends on the conditions of demand in the two countries. The greater the German demand for wheat, the more likely is it that the terms will be such as to yield Germany less wheat in exchange for her linen; the rate will be nearer 15 than 10. On the same reasoning, the less the elasticity of the German demand for wheat—the more her takings of wheat will persist with little abatement, notwithstanding a rising price of wheat—the more likely it is that the terms of trade will be disadvantageous to her. In the United States the moving forces are the same: the less her demand for linen, and the less easily that demand is stimulated by falling price, the more the terms will be to her advantage.⁰⁰⁵ In the second case supposed above, where the German demand for wheat has increased, the Germans, as the price of wheat begins to rise, will gradually lessen their purchases. But the extent of the check to their purchases will depend on the elasticity of their demand. The more persistent is that demand, the less it is checked by rising price, the greater will be the alteration of the terms of trade to Germany's disadvantage. The flow of specie out of Germany will then continue longer than it would if demand were less persistent (i.e. less inelastic), and the fall in [32] German prices and in money incomes will go on until the eventual limit is approached when there is no longer any gain at all to Germany from the exchange. In the United States, on the other hand, the extent of the readjustment will be affected by the state of her demand for linen. If that demand is elastic—if her people buy more linen quickly as the price begins to fall—the movement of specie into the United States will be less great than it would be if the demand were inelastic. The rise in prices and in money incomes will be less, the alteration in the barter terms of trade less markedly to the American advantage.

Something must be added to this. It is not merely the character of the American demand for German goods that has to be considered. Regard must be had also to the demand schedules of the Americans for their own product, of the Germans for theirs. When German linen falls in price, the Americans, while tempted to buy more linen, must consider the fact that in order to do so they must dispense with some wheat which they have been consuming. And the Germans on their part, when American wheat rises in price, have to consider that their payment of an additional price for the wheat necessarily involves a diminution in the amount they can spend on their own linen.

In other words, the supposition of the preceding paragraphs tacitly included the assumption that the Germans did experience this sort of double change in their demand schedules. When the German demand for wheat increases, as was assumed above, that very change necessarily implies that the German demand for linen is less insistent than before. They care more for wheat and less for linen.

Analogous, tho not quite the same, is the position of the Americans. They are offered more of linen than before for a given quantity of wheat, and have to decide whether they will take more linen and consume less of their own wheat. The character or intensity of their demand for the two articles is not supposed to have altered. It is only that, with demand schedules unchanged, they are called on to use less of wheat and to buy more of linen. While the outcome depends in both countries on the double aspect [33] of demand,—on the relation in each country between the conditions of demand both for the domestic product and for the foreign product—those conditions of relative demand have been supposed to alter in the one country, but not in the other.

In all this, be it noted, it is the changes in prices and in money incomes which serve to bring about the eventual results. Under the modified conditions brought about by an increase in the German demand for wheat, the concrete form in which advantage comes to the United States is that money incomes are higher, prices of imported goods (linen) are lower. And for the Germans the concrete form in which the disadvantage to them appears is that money incomes are lower, prices of imported goods (wheat) are higher. The quicker the flow of specie begins and the more prompt its influence on prices and incomes, the sooner will the readjustment work itself out. The whole chain of operations depends on monetary movements and monetary influences; a fact which cannot be too strongly emphasized when it comes to a testing or verification of the whole series of propositions presented in their simplest theoretic formulation.

[34] IN the preceding chapters the subject has been treated as if each and every commodity were within the range of international trade; or rather, as if there existed no other commodity than those within its range. Wheat and linen (or copper and linen) have alone been considered. Assuming that the conditions found for these are representative—that other commodities are in the same general situation as they—we reach the conclusion that some things, such as wheat, will be produced solely in the United States, while others, such as linen, will be produced solely in Germany. The price of each article will be the same thruout the trading areas (barring of course the differences that may result from cost of transportation). The purchasing power of money in terms of goods will be the same in the two countries. Money wages, however, will not be the same. In the case selected as illustrative they were found to be higher in the United States. The extent to which they may be higher—the maximum divergence—depends on the extent of the advantage possessed by the United States in the commodity which she exports. Within this maximum, the actual excess of American wages depends on the play of demand between the two bodies of consumers. The fundamental features of the situation, so far as its analysis has been carried hitherto, are that prices are the same in the two countries, while money wages vary.

But the supposition that all goods come within the range of international trade is not at all in accord with the facts. The scope of international trade is by no means all-embracing. So far from its being the case that each and every article is made solely [35] in one country and thence sent out to others—that all enter into foreign trade—it is more generally true that the goods made in a given country are sold and used in that country only. The conditions of international trade affect not the whole of a country's trade but only a minor part of it. We must distinguish between the international and the domestic goods: those, on the one hand, which are the objects of import and export trade and are the same in price thruout the international field; and those, on the other hand, which are not imported or exported at all, and do not necessarily have the same price in one country as in another. What can be said of the prices of the domestic commodities?

The essence of the answer to this question can be indicated by a simple illustration. Consider again the figures of the last case in the preceding chapter.

Wheat and linen are international commodities, each produced solely in the country having the comparative advantage for it. Suppose now we have bricks, so bulky in proportion to value that cost of transportation is prohibitive. They cannot be shipped from country to country, but are produced in each country, and are sold in each quite independently. Wages being higher in the United States, most persons would say that bricks also must there be higher in price. But this is not at all certain to be the case; the converse is just as possible. Suppose that

We suppose, that is, the effectiveness of American labor in brick-making to be high as compared with the effectiveness of German. labor in that industry. The United States has the same advantage over Germany in brick-making as she has in her export industry [36] (wheat); as regards both wheat and bricks the same labor produces twice as much in the United States as in Germany. The 10 days' labor produce 2000 bricks in the United States and but 1000 in Germany. Then, altho wages are higher in the United States, bricks are actually lower in price.

If now we change our supposition by assuming that the effectiveness of labor in the United States is not double that in Germany, but only one and one-half times as great,—if the 10 days' American labor produce 1500,—the American price of brick will be $1.00 per 100, identically the same as the German price. And if we change still further by assuming that the effectiveness of labor is the same in the two countries,—if in the United States as well as in Germany the 10 days produce 1000 bricks—the price will be higher in' the United States; the American price will be $1.50 and the German $1.00.

In other words, the prices of domestic goods are not necessarily higher in a country of higher money wages. They will be higher only if the effectiveness of labor is not higher in the purely domestic field. If the effectiveness of labor is positively higher than it is for the same articles in foreign countries, domestic prices may be as low in those countries or may be lower. High wages and high prices do not go together, either as regards international commodities or domestic commodities.

We may proceed now to indicate summarily what determines the range of money wages in a given country and what determines that of commodity wages.

Money wages, it is seen, are high in a country which has advantageous terms of international trade, which carries on trade with other countries in such way as to secure large gains from the trade,—favorable barter terms of trade. The main factors on which these gains depend have been sufficiently indicated: an outstanding comparative advantage, and the play of demand in the terms of trade. High money wages and incomes are the indication of favorable terms of trade, and constitute the mechanism by which the gains are secured. The countries having these favorable conditions realize them concretely by buying with their larger [37] money incomes foreign commodities which are cheaper than they would be if produced at home.

Regard, however, must be had also to what may be called the absolute effectiveness of labor. On this something has already been said, in our consideration of the nature and consequences of absolute differences in cost; and more will be said when we come to the competition of two (or more) countries when exporting the same goods to a third country.⁰⁰⁶ The bearing of absolute effectiveness can be readily indicated. Referring to our illustrative case, we might modify the figures by supposing that in the United States five days of labor, not ten, were required for producing the stated quantities of linen and wheat. We should then have the following:

The United States here has the same comparative advantage as before, but a greater absolute advantage in both commodities—an even greater all-around effectiveness of labor than was before assumed. Money wages in the United States are correspondingly higher—twice as high as in the original supposition. But the domestic supply prices remain as before; and so it is as regards the sharing of gains ascribable to the barter terms of trade.

The general level of prices is not necessarily higher in the country having the more effective labor, the more favorable terms, the larger gains from international trade. True, prices of international goods will be at the same level (still barring cost of transportation) in the favored countries and in those not favored. But prices of domestic goods obey laws of their own. Some of them may be higher in price than abroad, some may be lower; and the general level of domestic prices may therefore be higher or lower. So far as the effectiveness of labor in producing domestic goods is great (great, that is, in comparison with that of labor applied in other [38] countries to the same goods) they will tend to be lower in price. Conversely they will tend to be higher if the effectiveness of labor in producing them is small.

Concerning real wages also—wages in terms of commodities—no general rule can be laid down. Most persons would say that the people of the country where high wages prevail are more prosperous than those of countries with low wages. They may be or may not be. They are indeed better off as purchasers and consumers of international commodities; these being at the same prices everywhere. As regards domestic commodities, on which much the larger part of their money income is spent, they may be better off or may not be. It depends on the prices of these, which depend in turn, as we have just seen, on the effectiveness of labor in making them. The prices of domestic commodities may be higher than in other countries, and the people of the high-money-wages country, tho prosperous as purchasers of foreign goods, may be so much less prosperous in their domestic transactions that the net balance may be against them: their commodity wages may be less than in countries with low money wages.

I have just referred to the common but mistaken impression that a country of high money wages is necessarily more prosperous than one with low money wages. There are other common impressions even more widespread and more unqualifiedly wrong. Perhaps most familiar and most unfounded of all is the belief that complete freedom of trade would bring about an equalization of money wages the world over. It is a belief held especially in countries of high wages like the United States, and it goes with—indeed, is a part of—the most persuasive argument in favor of a policy of tariff protection. It seems plain as a pikestaff to the average person—to the average employer not less than to the average workman—that the country in which money wages are low can undersell the country paying high money wages; and that if the two compete without restriction, wages must become the same in both. The reasoning of the preceding chapters shows that there is no such tendency to equalization. Countries with high money wages trade with those of low money wages, to the advantage [39] of both, and with permanent maintenance of the divergences in wages. The reasoning is simple enough. Stripped to the essentials, as it here has been, it can be followed with ease; and in the sequel it will be shown to be no less convincing when elaborated, qualified, illustrated, and verified from manifold facts in the world of affairs. Yet to most persons it seems perplexing and anomalous, and remains so even tho the main lines are followed and accepted on the first summary presentation. In the field of economics, as in every intellectual field, fundamental principles, however simple, are not really understood until they are applied, repeated, turned over, gradually worked into the full intellectual equipment of the recipient. Elaboration, manifold testing, verification, are necessary not only for the refinement and accuracy of the principles themselves, but also for their assimilation.

Another notion, equally erroneous, relates to international differences in commodity wages. With the belief that unfettered trade between nations must lead to an equalization of money wages goes naturally the belief that it will lead to an equalization of "the standard of living"; and more particularly will bring a lowering of the standard of the prosperous countries to that of the less prosperous. People do not often distinguish with any care between money wages and commodity wages; they are apt to apply the same impressions and fears to both without discrimination. So far as they do distinguish, they fear equalization and lowering quite as much with respect to the standard of living in terms of commodities as to wages in terms of money. After what has been said of domestic prices and commodity incomes, it is superfluous to dwell on the point. Differences in commodity incomes as well as differences in money incomes may persist under complete freedom of trade. The causes of the possible persistence of the differences, even with free exchange between countries, are no less clear than they are with regard to money wages.

In general, then, we may say that high wages and incomes on the one hand, and high prices of goods on the other, do not go together. They do not go together as regards international goods; those are the same in price between countries, while money [40] wages vary. They do not go together as regards domestic goods; these vary from country to country, but do not necessarily vary in accord with money wages.

Nevertheless, there is a sense in which high wages and high prices do go together. Money wages and domestic prices run parallel; changes in money wages tend to accompany changes in domestic prices. If anything should occur which served to raise money wages—for example, altered and more favorable terms of international trade—a corresponding change would take place in domestic prices: they would rise to the same extent. Import prices, under the action of this factor, would not rise; they would fall. But domestic prices would adjust themselves to the higher level brought about by the new international conditions. Money wages would rise first in the export industries; the rise would then spread; eventually prices of goods and money wages in the purely domestic industries would be such as to render them as attractive as the export industries. This assumes, of course, that other things remain the same; that, for example, the technical methods of production remain unchanged—that no inventions or improvements are made which serve to increase the effectiveness of labor in the domestic industries. Such changes may operate to lower the price of a domestic article or series of articles at the same time in which international conditions are tending to lower them. We should then have the familiar case of interacting causes, in which the effect of the particular cause under inquiry is modified but not wiped out by others. Setting aside qualifications of this sort, which obviously do not affect the essentials of the conclusions, we may say that the relations of money wages and domestic prices, once they are established, remain the same. Domestic prices in the United States may be higher or lower than domestic prices in Germany; if higher, then a rise in United States money wages will carry them still higher; if lower, such a rise will make them less low.

It has already been intimated—to return to the main argument of this chapter—that the distinction between domestic and international commodities is an important one. How important [41] it is, will appear more fully as we proceed. It has been singularly neglected in the exposition of the theory of international trade and of its application. Most writers have dealt with the subject as if wages and prices not only moved together, but were necessarily at similar levels: as if high prices all around must go with high wages, and low prices all around with low wages. Commonly they speak as if advantageous terms of international trade must mean that the value of money is low all around, and prices high all around; not only that there are high money wages, but that all domestic goods, all services, all lands, houses, and lodgings are correspondingly high. Not so; the negative may be insisted on once more. And since the domestic trade of every country quite outweighs its international trade, and the portion of its national dividend that comes from its purely domestic trade is the greater, it follows that the negative is no less important in its practical applications than in its theoretical significance.

Altho the present chapter, like all in Part I, is concerned mainly with a theoretical formulation, most matters of verification and illustration being postponed to Part II, a word may be said here on the extent to which practical application can be made of the distinction between domestic and international prices. The needed statistical information too often is sadly lacking. In but few cases have we price records which separate the goods that enter into foreign commerce from those that come on the domestic market only. Most price data are prepared indiscriminately for any and all commodities, and most index numbers refer to one general (and for our purposes often confusing) price level. Some of the most interesting and significant points in the theory of foreign trade are difficult to verify—cannot readily be subjected to the test of conformity to the facts of the case—because we do not possess the data in suitable arrangement and classification.

It is to be observed, however, that in one important respect we are not entirely bare of the information we need; namely, on the rates of money wages. Here there are, at least for some countries and for recent times, instructive figures. They are more than instructive; they bear on the heart of the matter. The fundamental [42] thing in the movement of domestic prices, as distinct from international, is the movement of money incomes; and among these, again, the basic are the wages of manual labor, such as are usually registered in the index numbers of wages. Here is the item that is most significant and most easy to interpret and follow. Changes in money wages, so far as peculiar to a given country—so far as due, that is, to causes affecting it alone, and not a reflex of a general movement appearing throughout the commercial world—are at once the first indicator of changes on other domestic "prices", and also the effective mechanism through which the greater or less gain from international trade is transmitted to the several peoples. It is this which should especially be watched, and it is this, fortunately, which the statistical material, when available at all, is most likely to lay bare.

[43] IN the present chapter, a qualification will be introduced. It is not so much a correction as an elaboration; another of the steps which are needed in order to bring the theoretic analysis more nearly into accord with the facts of trade. It is one, moreover, to which virtually no attention has been given in the literature of the subject.

In the reasoning of Ricardo and Mill, it was almost always assumed that within a country commodities exchanged in proportion to the quantities of labor necessary to produce them. Goods exchanged par for par in terms of labor. In our own very first suppositions—those of trade conducted under barter—it was assumed that if in the United States 10 days' labor produced 20 of wheat and 20 of linen, wheat and linen would of course exchange within the United States at the rate of 1 wheat for 1 linen. And in Germany, if 10 days produced 10 of wheat and 15 of linen, 1 of wheat would exchange for 1-1/2 of linen. It was the very fact that within each country the relative values of goods depended on labor expended which led to the possibility of trade between the countries, and to the special problem of the terms on which they might barter their products. Similarly, when exchange thru the medium of money came to be taken up, the prices of goods within each country were supposed to depend on the labor given to making them—on their cost of production in the sense of labor applied. If 10 days' labor produced in the United States 20 wheat and 20 linen, wheat and linen would sell for the same price; with American wages at $1.50, each would sell for $0.75. If 10 days produced in Germany 10 wheat and 15 linen, wheat and linen would not sell for the same price; with German wages at [44] $1.00, wheat would sell for $1.00 and linen for $0.66-2/3. In other words, wages were assumed to be not, indeed, on the same level in the two countries, but at one and the same level in all industries thruout the United States and at one and the same level in all industries thruout Germany.

The familiar fact, however, is that there is no uniformity of wages within any country. There are differences within each country as well as differences between countries. The workmen who produce wheat in Germany may receive lower wages than those producing linen in Germany; and the wheat laborers in the United States may be in a position of similar disadvantage. Obviously such differences could not persist if there were perfect freedom of movement from occupation to occupation within each country; just as the differences of commodity income and of substantial prosperity between countries could not persist if there were perfect international freedom of movement. To designate the actual situation within countries it will be convenient to use Cairnes's phrase "non-competing groups." The workers in the several occupations (or groups of occupations) may be said to be in groups which do not completely compete one with another. There are persistent obstacles to transfer from one group to another, and therefore persistent differences of wages, not smoothed out by the movement of men from the lower-paid groups to the higher.

Given this sort of situation, it follows that the prices of goods are not in accord with the quantities of labor devoted to producing them. Even tho wheat and linen be produced in the United States with the same amount of labor, the two articles will not sell for the same price if the wheat producers get lower wages. On the other hand, two articles may sell for the same price even tho produced with different amounts of labor. Wheat and linen may be produced by different amounts of labor in Germany; yet, if the rates of wages are inverse to the labor amounts—higher where the days of labor are few, lower where they are many—wheat and linen will sell for the same price. International trade, however, like domestic trade, is proximately a matter of money sale [45] and purchase. Goods do not exchange directly for goods; they are sold for money and bought for money. The immediate actuating force is always that of the individual transaction—the sale of goods to advantage; and this means sale at a profit. How then can we have any assurance that such conclusions as were deduced in the preceding chapters concerning the influence of comparative labor costs have validity for the actual world? Goods are not bartered—wheat for linen—between countries. They are sold by individuals for cash. Their sale depends on prices; and prices are not necessarily, perhaps not usually, determined by quantities of labor given to producing the goods. How modify, adapt, reconcile our analysis of international trade to these plain facts ?

Let us revert to the case considered in the preceding chapter. The figures with which we there began, it will be remembered, were as follows:

The money cost of production of wheat is lower in the United States than in Germany; that of linen is lower in Germany. Trade takes place, the United States sending wheat, Germany sending linen. We still treat the wages outlay—that which the business world designates as "labor cost"—as if it were the sole item in supply price; return to capital is left for subsequent treatment.

Suppose now, that the German wheat laborers get as wages not $1.00 a day but only $0.66-2/3. Suppose them to be, among the Germans, in a non-competing group, unfavorably situated, receiving less wages than obtained in other industries, but unable to betake themselves to the more prosperous group and therefore Permanently in receipt of the lower pay. For the present, accept differences of this kind, whether in Germany or in the United States, as simply existent, disregarding the question how they [46] come to exist, and how far they may be related to international trade, in the last analysis, as effects rather than as causes. These aspects of the problem will be considered at the close of the present chapter. Taking the modified wages figures, then, we have further modifications in the other figures thus:

The supply price of wheat in Germany has fallen to $0.66-2/3. The labor in that occupation is comparatively ineffective, yet its price to the employing capitalist is low. Wheat can be sold at a comparatively low price, even tho it requires comparatively much labor to produce.

Observe further that both wheat and linen are now lower in price in Germany than in the United States. Both will be sold indiscriminately in the United States by German exporters. On the other hand no commodity can move from the United States to Germany. Specie will flow to Germany and prices will rise there. Prices will fall in the United States. The rise in Germany and the fall in the United States will go on until both wheat and linen sell for the same prices in the two countries.

The resulting situation will be of the following sort:

The movement of specie will then cease, and all movement will cease. There will be no trade between the countries. Each will go its way regardless of the other. The case will be the same in its outcome as that of equal differences in costs.

What this signifies evidently is that the lower wages for German wheat growers have the same sort of effect as would a higher effectiveness of their labor. In terms of labor-cost, of effectiveness [47] of labor, Germany has a comparative disadvantage in producing wheat. But this is offset, so far as the supply price of wheat is concerned, by the specially lower wages. In the market it is all one whether there be higher effectiveness of labor or lower rate of wages. Wheat would sell at a lower price ($0.66f) if the labor were as effective as German labor is in linen; it sells at that same lower price if the labor is obtainable at the lower rate of pay. International trade, to repeat, is governed proximately by prices; and the ruling prices, under this supposition of a non-competing group, are such that wheat is not sent from the United States to Germany even tho the United States has unmistakably a comparative advantage for producing it.

Now make a further supposition. Suppose that not only in Germany, but in the United States as well, the wheat laborers are in a low-lying group; that these Americans, like their German fellows, receive lower wages than obtain in other occupations, and therefore lower wages than the linen workers. Suppose that while wages in the United States are in general $1.50 a day, the wheat workers get no more than $1.00 a day. The German wheat workers, in their turn, receive only $0.66f a day, as against a ruling German rate of $1.00 a day. Then our figures must be modified as follows:

Prices are now such that commodities move both ways. The American supply price of wheat ($0.50) is now lower than the German supply price ($0.66f). Similarly German linen at $0.66| is lower in supply price than American linen ($0.75). Trade between the two countries takes place as it would if the differences in wages within each of them did not exist—as if there were uniformity of wages in each and no non-competing groups at all.

The general conclusion thus indicated is that the existence of non-competing groups within a country affects international trade [48] only so far as the situation thus engendered is peculiar to that country. If the groups are in the same relative positions in the exchanging countries as regards wages—if the hierarchy, so to speak, is arranged on the same plan in each—trade takes place exactly as if it were governed by the strict and simple principle of comparative costs. If the rate of wages in a given occupation is particularly low in one country, this circumstance will affect international trade exactly as would a high effectiveness of labor in that country. But if in other countries also the same occupation has a particularly low rate of wages, international trade will not be affected. The coefficient to be allowed for will be the same all around, and no special influence on trade between the countries will be felt. Trade will develop as it would if prices within each country were governed by labor costs alone.

For further illustration, let us turn to a variant of the previous case. Starting with a situation in which, so far as labor costs go, exchange cannot be expected to arise, introduce the complication of non-competing groups and observe how under the changed conditions exchange becomes possible and advantageous.

The case, it will be seen, is one of equal differences in cost. The effectiveness of labor in the United States is twice as great thruout as in Germany. Money wages in the United States are adjusted to this relation and are twice as high. Wheat is at the same price in the United States as in Germany; linen also at the same price in the two countries. Wages in each country are uniform—that is, are the same in wheat-growing as in linen-making. There are no non-competing groups. Prices are in accord with the respective quantities of labor. Germans and Americans go their way regardless of each others' doings.

Suppose now that German linen wages are not $1.00 but $0.75. The linen workers are in a low-lying non-competing group; their [49] wages are less than prevail in other German industries. Our figures then are modified thus:

Wheat is still at the same price in both countries. But linen is now cheaper in Germany, and moves from Germany to the United States. At first specie alone moves from the United States to Germany. As prices fall in the United States and rise in Germany, wheat becomes cheaper in the former and dearer in the latter. Wages fall in the United States, rise in Germany. Equilibrium will be reached under conditions somewhat like these:

This particular relation of prices and of money wages would be reached (wages in the United States $1.80, wages in Germany generally $1.10), as need not again be explained if the demand for linen in the United States and the demand for wheat in Germany were such that the money value of the wheat sent from the United States exactly equalled the money value of the linen sent from Germany. The general outcome would plainly be that grade developed precisely as if the Germans had a comparative labor advantage—a comparative effectiveness of labor—in making linen. The conditions of labor cost are such that if these were the only governing factors, no trade between the two countries would develop. But the exceptionally low wages of the German linen workers cause Germany to have the equivalent of a comparative advantage. The case is the converse of that just considered.

Now push the matter a step still further. Tho we may conceive of non-competing groups as separate and distinct, it never happens [50]—virtually never—that a given commodity is produced solely by laborers of one group only. The usual situation, when once the division of labor has been considerably developed, is that a commodity is made by a combination of laborers belonging to different groups. The several laborers whose work serves to turn out linen, for. example, will probably not be all in the same stratum; some will be well-paid, such as the mechanics who make and repair the machinery, others will be unskilled operatives who tend and operate it. The combinations are various in the different industries, various in different stages of industrial development» various in different countries. They are likely to be less heterogeneous (to refer again to the example of linen) where there is a household handicraft industry, such as long persisted in Germany, than where there is a highly developed factory system, such as alone is to be found in the American textile industries. The combinations are likely to be more heterogeneous and elaborate in manufactures than in agriculture; more so in countries industrially advanced like England or Switzerland than in those industrially backward like Spain or Portugal. But in every case, if account is taken of all the labor involved in producing a given article—of the labor given to the raw material, of that fashioning it, of that transporting and marketing it—some combination of different grades of labor will be found. The theory of international trade must be adjusted to this all-pervading heterogeneity. For illustration of the working of this factor, return to a case of comparative costs such as was considered in the initial stages of our analysis.

The case is one in which the United States has a comparative advantage in wheat—a superior advantage. Germany has a comparative advantage in linen—an inferior disadvantage. Under barter, the two would obviously find it advantageous to exchange American wheat for German linen. Under a money [51] regime, with no non-competing groups and with prices adjusted in accord with labor costs in each country, the same result as obviously would ensue.

Suppose now something like the usual industrial situation: not merely non-competing groups, but also laborers from different groups combined in the making of any one article. Suppose that in the United States there are some groups whose established pay is $1.50 a day, others whose established pay is $1.00 a day. In Germany there are groups whose established pay is $1.00, others with $ 0.66-2/3. In each country higher-paid and lower-paid laborers are joined in the making of linen, and are joined also in the making of wheat. For simplicity, suppose that in each industry one-half of the laborers thus combined are from the upper stratum, one-half are from the lower stratum. Then we have the following:

Observe the domestic supply prices of the two articles. Wheat is cheaper in the United States, linen is cheaper in Germany. Wheat goes from the United States to Germany, linen from Germany to the United States. Precisely the same sort of trade takes place as would be found if there were no non-competing groups.

The general proposition to which this leads is simple enough, and indeed hardly needs to be brought out by figures. If the combinations of several sorts of labor, paid at varying rates, are the same in the two countries (the hierarchy of the groups being also the same), trade between them takes place exactly as if there were no internal differentiation at all—as it would if there were no non-competing groups. It is only a difference in the arrangement of the industrial hierarchy, not the hierarchy itself, that has effect on international trade. To change the simile, given

[52] the same industrial stratification thruout, the fact of stratification is of no consequence; the several layers are related to each other as if they were a pair of homogeneous structures

Such figures could be easily varied further, and the same principles further illustrated. We may suppose the relation between the rates of pay in different groups not to be the same in the two countries. That is, suppose non competing groups in each country, but with differences between the groups not the same in both. If the lower paid laborers in the United States (the unskilled, say) receive not only lower wages than those belonging to the higher groups, but wages farther down in the United States scale than is the case with corresponding sorts of laborers in Germany, then the commodities for whose production they are combined with the others will be affected in price exactly as if that sort of labor were especially effective in the United States. These commodities will be relatively cheap and will tend to be exported. Again, if it happens not only that a given group of laborers gets an unusual rate of pay, but also that a large proportion of this sort of labor is needed for producing a given commodity, the result will be accentuated. Assume, for example, that skilled workers of a given kind are to be had in Germany at a premium or differential over other workers which is not so high as the premium for the same skilled workers in the United States; assume further that the technical processes of an industry require a proportion of such workers larger than is needed in other industries; then the products of that industry will be particularly low in price in Germany, even tho not made with labor having any particular (comparative) effectiveness. They will tend to be exported; they may be exported even tho the labor lack something in comparative effectiveness.

There is more to be said, however. What causes the differences of wages within a country ? What determines the relative positions of the non-competing groups? The underlying forces may be solely of domestic origin and effect; then they are to be conceived as operating on international trade as separate and independent [53] factors. Or they may be themselves partly of international range; then the conditions of international trade themselves operate as cause, and the domestic and international factors become mutually dependent.

The underlying forces are solely domestic if social or industrial stratification within a country rests on deep-rooted differences in the standards of living of its several groups. This hypothesis has been stated with admirable precision by Marshall:

"Suppose that society is divided into a number of horizontal grades, each of which is recruited from the children of its own members; and each of which has its own standard of comfort, and increases in numbers rapidly when the earnings to be got in it rise above, and shrinks rapidly when they fall below that standard. Suppose, then, that parents can bring up their children to any trade in their own grade, but cannot easily raise them above it and will not consent to sink them below it.

"On these suppositions the normal wage in any trade is that which is sufficient to enable a labourer, who has normal regularity of employment, to support himself and a family of normal size according to the standard of comfort that is normal in the grade to which his trade belongs; it is not dependent on demand except to this extent, that if there were no demand for the labour of the trade at that wage the trade would not exist. In other words, the normal wage represents the expenses of production of the labour according to the ruling standard of comfort."⁰⁰⁷

In such case the relations of the several groups are settled by causes quite independent of international trade. They would persist if there were no such trade at all, and would be no more potent and no less if such trade took place on a great scale. The groups get their several rates of remuneration because of differences in the conditions of supply for the several kinds of labor and service, not because of the quasi-fortuitous impact of demand. They operate as causes of price (the price of a particular kind of labor); they are not the results of the price of goods. They are [54] purely domestic, in the sense that they rest on the standards of living in the groups, which are the outcome of historical and social forces in the given country.

But the causal sequence—still speaking of the domestic situation by itself—may not be so simple. It may be that the rates of pay in the several groups are settled by the mere conditions of demand; or, if not absolutely settled, affected or modified by those conditions for periods so long that they cannot be ignored even in inquiries that disregard short time phenomena. As is well known to the reader conversant with the history of doctrine on this topic, Cairnes treated the relations between non-competing groups as dependent solely on demand. The principle of the play of reciprocal demand, to which Mill had turned for the explanation of the barter terms in international trade, was applied by Cairnes to the explanation of exchange between groups within a country. That group whose services (goods) were much wanted by other groups, and which itself wanted little of the services of other groups, was able to secure the greatest advantage from the exchange; it had the highest scale of earnings. Cairnes did not proceed to the apparently obvious corollary that numbers played a part in those relations. Any group whose numbers are small, confronted by another whose numbers are larger and exchanging products with that other, is likely to secure advantageous terms in the play of reciprocal demand. And the question of numbers raises that of the increase of numbers—the marriage-rate and birth-rate, and the standard of living. But this series of questions, to repeat, was never raised by Cairnes; the problem was treated by him as one solely of demand. And it may well be that we are not in a position to say much of the other side of the problem—the conditions of supply. The standard of living, as between different groups, can hardly be said to be well defined, still less to be well settled. Of necessity it acts very slowly in its effect on numbers. During the course of the period which must elapse before it can operate with effect—one or two generations—the impact of demand may shift. The relative rates of pay may shift accordingly, rising here, falling there; and the effects of [55] a shift may endure so long that the standards themselves may change. There is as much evidence to show shift in the standards of living of different classes as there is to show fixity. It may fairly be maintained that when we pass beyond the forces of demand (which are in any case determinant only over many years) and try to examine the forces of supply, we do reach not a domain of fixity but one of constant flux.

If this be the just view, it follows that the impact of reciprocal international demand is not separate from that of reciprocal domestic demand, but merges with it and becomes part of one combined force. For example, we have supposed, in a previous illustrative case, that the linen makers of Germany are in a lower group, less well paid than other German workers. While German wages in wheat are $1.00, they are but $0.75 in linen. Why the difference ? Is it due to settled standards of living in the several groups ? To supply or to the play of demand ? We may hesitate to go further than say that the numbers of persons in the other groups and the keenness of their demand for linen, compared to the number of the linen-workers and the keenness of their demand for the other products, have combined to bring about by a quasi-mechanical process the stated differences in the wages of the German exchanging groups. If this then be regarded as the initial situation—that established in Germany when isolated—and if we suppose her thereafter to be confronted with a demand for linen from the United States, this new demand for German linen is added to the former demand from the German workers themselves. The play of demand is altered to the advantage of the linen workers. The relations between the groups within Germany become different; the linen workers get higher prices for linen and higher wages for themselves.

How important in practice is the general train of reasoning followed in this chapter? Are we to conclude that the more simple analysis with which we started, resting on the assumptions of homogeneity in labor groups and uniformity in wages, becomes quite inapplicable where there are heterogeneous social and industrial conditions and wide diversities of wages in any one country?

[56] The answer, as already indicated, depends not so much on the existence of non-competing groups in the several countries as on the similarity or dissimilarity of their make-up. Their bearing on international trade depends on whether they are of the same sort or of different sorts in the trading countries. Now, in the occidental countries—those of advanced civilization in the Western world—as a rule the stratification of industrial groups proceeds on the same lines. And it is between these countries that the principle of comparative costs is presumably of greatest importance. Since differences of climatic and physiographic character are less wide, divergences of absolute costs are less common and less great, and the limits within which the terms of exchange are confined not so far apart. And in the Western countries, to repeat, we find roughly the same social and industrial layers. The unskilled, by far the most numerous, get the lowest wages; the mechanics and well-trained stand distinctly higher; and so upward. This being the case, the differences in money costs between the countries are mainly determined by differences in labor costs; even tho within each country this factor may be profoundly modified.

Further: that combined influence which domestic and international demand may exercise on the position and prosperity of a given non-competing group is not of so great importance in practice as it is for the completeness and consistency of theoretic analysis. It is less important because the demand from abroad for any set of commodities, and thereby for the services of a particular grade of labor applied to making those commodities, is rarely so dominant as to change those relations between grades which would obtain within the country in any case. The lines of social and industrial stratification in a country are determined chiefly by the conditions that prevail within its own limits—by the numbers in the several groups and their demands for each others' services, and in some uncertain degree by their different standards of living. An added impact of demand from a foreign country will rarely change the relative rates of wages which have come about from the domestic factors. The social stratification [57] that results from the domestic conditions is well established and seems to be deeply rooted; and it is not likely that international trade will impinge on it with such special effect on a particular grade as to warp it noticeably.

The case is somewhat different as regards the train of causation running the other way. While international trade is not likely to modify the alignment of grades within a country, peculiarities in that alignment may affect international trade. I will call attention to one or two instances in which this sort of influence seems to have appeared, departing for the moment from the general plan of this book, under which illustration and verification have been relegated to the later chapters.

The first illustration comes from the history and position of the chemical industry of Germany. I speak of the situation as it was before the war of 1914-18; what happened in Germany in the years immediately after the war is too confused for the illustration of the forces ordinarily at work in international trade. Before 1914, as is well known, chemical industries, and especially those yielding highly elaborated coal-tar products, were more successfully carried on in Germany than in any other country. Coal-tar dyes and drugs were supplied to England and the United States from Germany; the domestic output in these countries was negligible. Other countries also were supplied by German imports, tho not as preponderantly as the two English-speaking countries. The Germans evidently had some advantage in making these things. A comparative advantage? Certainly not one of a natural (physical) sort. It arose largely from the plenty and the especial cheapness of a particular kind of labor: that of chemists and of chemists' skilled assistants. Germany had a learned proletariat. The excellence and easy access of technological education, and the powerful social forces which attracted large numbers from the middle classes into the learned professions, brought about a large supply and a low remuneration of highly trained chemists. A similar excellence of intermediate education supplied to these officers a capable non-commissioned staff; (to use a military analogy) there was a supply of exact, careful [58] assistants and workmen, also paid at rates low in comparison to those of other countries. I will not say that this was the only factor that served to give Germany her unique position in the coal-tar industries. There were others, not least the marked faculty for elaborated organization which had developed during the latter years of the 19th century; a faculty that told with special effect in an industry like this—intricate, large in its scale of operation, yet not characterized by mass production. For the present purpose it is enough to note the influence of the labor situation. The special cheapness of the types of labor needed to an unusual degree in the industry served to give it a comparative advantage—that is, an advantage in the pecuniary terms which are decisive in the markets. And the advantage doubtless was not confined to the coal-tar and other chemical industries. It was probably generic. It appeared in scientific industries of other kinds, such as for example the making of optical instruments, surgical, instruments, laboratory apparatus. Not one industry only, but a considerable number of German industries similar in character were given a place of their own in international trade because of the special position in Germany of the grade of labor needed for their products.

Quite a different illustration, derived from the situation of a group lying not in the upper line of workers but in the lower, is to be found in the United States during the same period. A marked peculiarity of the American labor situation during the generation preceding the Great War was the comparatively low rate of pay' for the unskilled laborers. It was low, that is, in comparison with the pay of the upper stratum of the skilled laborers. While the pick and shovel man got more in the United States than in Europe, he did not get as much more above the European rate as did the American mechanic. The differential in favor of the mechanic was greater in the United States; the unskilled were relatively cheap, even tho not absolutely so, for the American employer. The cause is not far to seek. The enormous influx of immigrants maintained a great supply of unskilled labor and kept down its rate of pay. In the manufacturing industries of the [59] Southern States the utilization of a low-lying stratum of "poor whites" (not to mention the negroes) operated in the same way. The effect was to give an advantage to those industries, or those ways of conducting industries, in which the low-lying group of labor was used in large proportion. Industries of this type were accordingly in the same position in regard to international trade as if they had a comparative advantage; or if not so much as this, something to offset a lack of such advantage.

In the iron industry—that is, in the making of crude and half-finished 'iron and steel—the effect was of the former sort: the situation served to give a comparative advantage. The industry uses great masses of labor. The industry grew in the United States at an extraordinary pace between 1890 and 1915, and came to be an important industry of export. Here, too, the labor factor was not the only one; but it was an important one. It contributed to the remarkable overturn by which the United States, formerly an importer of iron and steel, became a great exporter of them.

In the textile industries an analogous development took place, but here not so much in the way of greater exports as of less imports; not so much the attainment of a clear comparative advantage as the elimination, in part or in whole, of a lack of superiority. The shift for the purposes of international trade was negative rather than positive. Those textile industries which could use unskilled labor for tending semi-automatic machinery for mass production found a plentiful and cheap supply at their command. Those for which still other conditions also were favorable, notably those manufacturing the cheap and medium grades of cotton fabrics, grew apace. Their position of indifference to foreign competition, almost if not quite attained even under the earlier conditions, was strengthened and consolidated by the cheapness of the routine labor. Textile industries of a different type, such as the silk and worsted manufactures, were enabled to attain a half-way position. For them the general conditions were less favorable; in order to hold their own against foreign competition, they needed a tariff prop much more than [60] did the leading branches of the cotton manufacture. But the utilization of cheap common labor enabled them, not indeed to hold their own without protective duties, but to get on with a less barrier than would otherwise have been called for. The effect was the same in kind as that on the cotton industry, but not so marked in degree.

These peculiarities in the American labor situation did not rest on permanent causes. They were due, as has already been said, primarily to the great inflow of immigrants during the period in question. The restrictive legislation of 1916 brought a complete change, one whose effects will ramify far and in many directions, but in no way more than in a new adjustment of the relative wages of skilled and unskilled laborers. The differential will become less pronounced in favor of the skilled as against the unskilled. The industries which have adjusted themselves to a large and relatively cheap supply of the unskilled will have to readjust their ways. So far as they are subject to competition from foreign industries, they will be in a less advantageous position than before. The relations between the wages of the two groups will probably come to be in the United States not different from those in England, in Germany, and in Australia. This particular source of comparative advantage (or of an offset to a comparative disadvantage) will grow less and less, and probably will in the end disappear.⁰⁰⁸

[61] STILL another factor, that of capital and interest on capital, will now be considered. We have seen that while the mere shift from cost in terms of labor to supply price in terms of money wages did not modify our conclusion, the consideration of differences in wages and of their influence on the money expenses of production did lead to significant modifications. So it will prove as regards interest on capital. Obviously this constitutes an item in the expenses of production; and it is one that has effects of its own. Yet these also are such as rather to modify the general conclusions reached on the simpler suppositions than to overturn them.

It is hardly necessary to remark that we need not consider separately such an item as the expense for materials—one that would bulk large in an accountant's schedule. It is familiar in economic doctrine that expense for raw materials may be resolved into previous expense for wages and interest. For the purposes both of international trade and of domestic trade we bring together in one sum total all the costs and expenses involved for a given article; not only those of the immediate producer, but those of the antecedent persons from whom he buys materials and supplies, and whom he recoups (with interest) for their expenses in the way of wages and still earlier materials. In the same way, when we considered the principle of comparative costs in its simplest aspect (disregarding money expenses and prices) we attended to all the labor involved in producing an article, not merely that of the last stages in its production. Raw materials, then, may be brushed aside, as involving an embodiment of previous labor and a recapitulation of previous wages and interest.

[62] Rent, let it be briefly noted, is also to be brushed aside. In accord with the commonly accepted procedure in economics, we shall treat it as merely a differential element. It stands for the differences in the expenses of production under varying natural conditions, and serves to equalize them. Some relations between rent and international trade deserve attention and will receive it in the next chapter. For the present we shall disregard them.

We proceed then to consider the influence on international trade of an interest charge as one among the expenses of production. It will be convenient for this phase of the inquiry to revert to the second of the three original cases: that, namely, in which there are equal differences in cost and in which international trade will not arise. It appeared that trade could not be expected to arise, if regard were paid solely to labor and to wages—to quantity of labor and to the wages of labor. How if the additional factor of an interest charge is introduced ?

Recall the former figures:

There are equal differences in cost; money wages are adjusted to those differences; the price of each article is the same in the two countries; no trade takes place.

If now we suppose a flat addition to be made, at the same rate, to the expenses of production all around, the possibilities of trade will be no greater. Add 10 per cent or 50 per cent (according as there is little expense, or much, in addition to the outlay for labor) to the wages bill in each case, to stand for interest. The figures then are all enhanced by the same amount, prices remain the same in the two countries, and no trade will arise. The mere circumstance that there is a return to capital leads to no modification of the analysis based on labor costs and wages alone.

Now change the situation by making the rates of interest not uniform thruout, but higher in one country than in the other.

[63] Suppose the rate to be twice as high in the United States as in Germany. The interest charge (not the same thing, of course, as the rate of interest) to be added to the expenses of production then becomes twice as high. Suppose the interest charge be 50 per cent of the wages bill in the United States, 25 per cent in Germany; then we have:

Both copper and linen are lower in price in Germany; both move to the United States; specie flows from the United States to Germany. Prices and money wages will fall in the United States, rise in Germany. Copper will rise in price in Germany until it is at the same price as in the United States; linen will rise similarly. After the redistribution of specie, the price of each article will be the same in the two countries, tho at a level somewhat higher all around than in the bare and simple situation first considered. Money wages will be readjusted, at rates somewhat lower than before in the United States, somewhat higher in Germany. The change in money wages signifies that with the higher interest charge in the United States the share of total national income which goes to the laborers is smaller there than in Germany. So far as concerns international trade, nothing happens except the temporary movement of goods one way and the redistribution of specie. Once this much is accomplished, the two countries have no trade connection; each goes its independent way. The mere fact that the interest charge is lower in the one than in the other does not cause the conditions for international trade to be different from what they were before.⁰⁰⁹

Now change the situation in still another way. Assume that one of the articles is produced with much capital, the other with little. Make the case extreme: suppose that one of them (copper) [64] is produced with the aid of very much capital, the other (linen) with none at all. Suppose that the American capital investment for copper is $100, and that during the period over which $20 is paid for wages, 10 per cent on $100, or the sum of $10 in all, is payable to capital for interest. In Germany suppose the copper situation to be of the same sort. Then the German capital investment for copper will be $66.66, because the same quantity of labor applied to making that capital, paid at the German rate and having the German effectiveness, will bring the investment to that sum, as against $100 in the United States. The German interest charge for copper, at 10 per cent, will then be $6.66. For linen, be it remembered, there is no interest charge. Then we have:

Both in the United States and in Germany copper is now higher in price than it was before the interest item was added. It was $0.66-2/3 in both countries before the interest charge was added; it is now $1.00 in both. Being the same in both, copper moves neither way. The price of linen remains unchanged, still at the old figure and still the same in either country. Trade between the countries still will not arise.

Suppose now, however, that there is an interest charge for copper (and still for copper only—none for linen), but not at the same rate in the two countries. Suppose the rate to be 5 per cent in the United States, while it remains at 10 per cent in Germany. Then we have:

The situation has changed. The price conditions are such that copper will move from the United States to Germany. Tho linen [65] is still at the same price in both countries, copper has become cheaper in the United States, and can be profitably exported to Germany. Linen at the outset will not move from Germany; specie will be sent to pay for the American copper. As prices and money incomes rise in the United States and fall in Germany, linen will become dearer in the United States and cheaper in Germany, and will begin to move.

The extent of the consequent readjustment of prices depends as need not again be explained, on the conditions of demand for copper in Germany, for linen in the United States. A possible outcome would be the following:

We now have the conditions under which both articles will move. Copper is cheaper in the United States, as before; but linen is now cheaper in Germany. International trade arises and rests on enduring conditions. It will go on indefinitely, to the advantage of both countries.

The general proposition to which this series of illustrative figures points is that interest on capital acts on international trade not in itself, but only in so far as it operates differently on different commodities. At the very start it is obvious that an interest charge, added uniformly to the expenses of production, brings no alteration in relative prices, since it acts equally on all commodities. Nor does an interest charge have an effect simply because it is at a different rate in the two countries—higher in one than in the other. If within each it acts uniformly thruout, it [66] leaves the relations between the countries undisturbed. Further : an interest charge does not alter conditions for trade, even tho it act on one commodity only, provided it acts on that commodity in the same way in both countries. The same is true (illustration in figures may be spared) if the interest charge, instead of being absent on one article while present on another, is merely greater on one than on another. If the difference is the same in both countries, international trade goes on in the same way as if this factor had not entered.

But the circumstance that the rate of interest is higher or lower in a country does have an effect when the needed capital equipment is greater for one commodity than for another. It bears more on those commodities which are made with much capital, making them relatively higher in price in the country where there is a higher rate of interest and a higher interest charge, lower in a country where there is a lower rate.

A low rate of return on capital, then, tends to give to a country a comparative advantage (i.e. the equivalent of one) for those goods which are made with much capital; these tend to be exported from it. A high rate of interest is correspondingly a handicap on the export of these same goods, a stimulus to their import. To put it in a more concrete way, a country in which capital has accumulated in large amounts and in which the investors are content with a low rate of return, tends to export articles which are made with much plant, and with raw materials which it takes time to produce and transport; whereas a country in which accumulation is smaller and the interest rate is higher, tends to import such articles. High or low interest does not in itself act as an independent factor; it exercises an influence of its own only so far as it enters to greater degree in one commodity than in another.

The conclusion is of essentially the same sort as that reached with regard to non-competing groups and differences of wages. So far as differences of wages are the same in two or more countries, and so far as goods are made in these countries with the same assortments (combinations) of different grades of labor, international trade remains as it would be in the absence of this complication. [67] Only so far as there is a peculiarity in the position of a particular laboring group; or so far as a commodity is produced in one country with a different labor group or assortment of groups from that which is utilized in another—so far only does this factor exert a modifying influence of its own. The investment of capital and the payment of interest on capital are to be regarded in the same way. No essential modification of the original analysis is called for, unless this factor in turn leads to price phenomena which are different for one commodity in a given country from those for the same commodity in another country. A higher or lower rate of interest, so far as it operates on a particular commodity with greater effect in country A than in country B, has its influence in causing the price of that commodity to be different, relatively to other commodities, within the country; and thereby only does it have an influence of its own on international trade.

The quantitative importance of the capital charge factor in international trade is probably not great. As the whole tenor of the preceding exposition indicates, the range of its influence is restricted to a special set of circumstances. Within that range, its influence is further limited by the absence of wide inequalities in the rate of return on capital. Interest, while it does vary somewhat from country to country, does not vary widely between the leading countries of western civilization; and it is in the trade between these, and in the competition between them for trade with other countries, that the interest factor is most likely to enter with its independent and special effects.

The analogy to non-competing groups and laborers may again be applied. Capital may be regarded, if one pleases—of that I shall say a word presently—as an independent factor, competing with labor so far as concerns contribution to the output. But it does not compete with labor in the sense that there can be any equalization of sacrifice between capital and labor; the two sacrifices ("abstinence" and work) being in their nature incommensurable.⁰¹¹ Marked differences in the rate of return on capital, persisting [68] indefinitely, are indeed quite conceivable, just as marked differences in the range between the wages of the several non-competing groups of laborers are quite conceivable; and there might be corresponding results of some quantitative moment in international trade. But since, as a matter of fact, the differences in the interest rate between countries are not considerable, we are justified in concluding that this element in the economic situation, like the element of persisting differences in wages to different workers, does not lead to a radical modification of our first conclusions.

There is more to be said, however, concerning the way in which the use of capital bears on international trade; there is another point of view. Tho the rate of return on capital may have no such marked influence as is often supposed, the fact that capital is used, and is used with far-reaching effects on the effectiveness of labor, has consequences in international trade which in turn are far-reaching.

Capital and labor are often referred to as agents which compete in production. This form of statement, useful for some purposes tho it is, does not describe with accuracy what happens when capital is made and used. Capital is itself made by labor; and the use of capital simply means the application of labor in another way—by an indirect and prolonged process. When a workman uses a tool in making a given article, the total labor given to produce the article includes not only what is done directly by the tool-user, but also a part of the labor of the tool-maker. If, for example, a tool is made by one man, and lasts just a year; and if another man then works with that tool during its year of life; then the resulting articles are produced by the labor of two men each working one year, or of one man working two years. Similarly, if 100 men work with machinery made by another 100 men; and if the machinery lasts 5 years; then in any one of the five years, the labor given to the resulting product is that of 120 men. This much is a commonplace in economic theory. It was clearly stated long ago by Ricardo,⁰¹² and was explained, elaborated, insisted on, by Bohm-Bawerk.

[69] The concrete way in which the element of previous labor is reckoned by the business world is through the charge for depreciation in cost accounts. If machinery lasts 5 years, a considerable item must figure in the expenses of production to make up for its depreciation; if it lasts 20 years, the allowance is less, but is still there. Something must always be set down on this score; unless indeed the machinery last forever.

Now, in the illustrative figures used in the present chapter, no allowance at all was made for any such item. In other words it was tacitly assumed that capital (machinery or what not) did last forever. On that assumption the only new element brought into the account by the introduction of capital is the returns on it; past labor and depreciation need not be considered. In the actual world, however, it must always be considered. When making up the complete summation of the labor given to an article, we must put down something for the labor of the past which has been given to making the tools or machinery. It will be much or little, according as the capital instruments last a short time or a long; much or little, according as depreciation bulks large or small in the accounts.

Our total for the expenses of production (referring now to one of the previous illustrative examples), as modified by the introduction of capital, might then be stated in some such form as this:

There is still more to be considered, however, than this revision of the method of figuring. The revised calculation (as the reader is likely to say) in itself adds nothing of moment. The number of days' labor for the given article, and the wages item in the expenses of production, became greater, and the figures are readjusted accordingly. What really signifies lies in quite another direction. The use of capital means not merely that an apportionment must be made (perhaps somewhat intricate) of the total labor given per [70] unit of output. It means also that the effectiveness of the labor per unit is increased. The use of good tools and machines enables the same product to be got with much less of current labor. The illustrative figures just given would imply on their face that with the use of capital the total expense (the supply price) per unit becomes higher. For closer verisimilitude, they should look something like this:⁰¹³

But further—and here we reach at last the point which is of importance for the theory of international trade—this reduction in the total labor applied, the increase in the effectiveness of labor, the lowering of cost in terms of labor and in terms of money, the whole train of modifications—is likely to take different shape in different countries. And different not only between countries, but between commodities. Some countries use tools and machines more readily and more effectively than others; some commodities are more amenable to the machine processes than others. Comparative advantages and disadvantages emerge.

These are advantages and disadvantages, be it remembered, arising from the relative effectiveness of the totals of the labor applied. They arise, not because the matter of return on capital is involved, but because a more complicated reckoning must be made of the effectiveness of labor. This fundamental fact is disguised by the business man's and the accountant's ways of reckoning. [71] Depreciation is treated precisely as is the outlay for materials and supplies. Such items are treated in the accounts as if they were quite separate from the wages bill and the "labor cost." For the purposes of the economist, however, and not least for the theory of international trade, they must all be reckoned in the labor account. Their significance for our purposes lies in the fact that the economic analysis of capital outlays points to differences in labor cost; to variations in this essential regard from country to country and from commodity to commodity.

A country that makes large use of tools, machines, plants, and uses them better than another country, has a comparative advantage in the production of the commodities turned out with the abundant use of capital. Such in general is the situation between the countries of advanced capitalistic development, Western Europe and the United States, when compared with the tropical and backward countries. As between the Western countries themselves, there are similar differences. England had a marked advantage of this kind for a considerable period, from the early stages of the Industrial Revolution in the 18th century through the first third of the 19th, perhaps the first half. England continued during that period to have a comparative advantage in making those articles to which the machine-using processes could be applied with most effect; all countries applied them more or less, but England applied them better. The United States attained a development of a similar kind by the middle of that century, Germany and Switzerland before its close. Gradually all the Western countries learned to apply the new labor-saving processes. Yet they did not all learn to apply them with equal effect. Differences persisted, and these had their effects on international trade.

A curious contrast appeared in the latter part of the 19th century between the situation in England and that in the United States. It serves to bring into clear relief the distinction between the two ways in which the use of capital affects international trade, according as it operates on the one hand to introduce the element of return on capital or on the other hand to increase the effectiveness of labor. In both countries the use of well-devised tools and [72] machines had been carried far—doubtless farther than in any other parts of the world. In both, the effectiveness of labor was made greater by the capital-using method of production. Probably the United States was somewhat in the van. Hence in those industries which were specially suitable for this method of production she had a comparative advantage. Doubtless she had in this regard an advantage in all industries, since her people thruout devised better tools than other peoples and used them better; but international trade was influenced only in so far as there was a peculiar—a comparative—advantage in some among her industries ; that is, in so far as United States used capital better in some directions than in others. England also used tools and machines with large effect, even tho not with all-around effect as great as the United States. In England, however, the other side of the capital factor entered, giving a comparative advantage in certain directions; namely, the return which had to be paid in the way of interest on the needed capital was lower. So far as all of her industries shared in the lower interest rate alike, no influence of international trade could emerge. But so far as the technical development of a particular industry called for large capital—a large amount of previous labor allied with a moderate amount of current labor—the lower return on the capital embodying the previous labor gave that industry a comparative advantage. The iron industry was typical: larger plant, larger outlay for materials, comparatively small current-labor account. Here a low interest charge had a greater effect on prices, and thru prices a greater effect on international trade, than in industries where capital charge was a less important item. The United States, on the other hand, at this later stage, had an advantage in those industries when the use of tools and implements made the effectiveness of labor especially great. England had an advantage in those industries where much capital was used and where the lower interest rate enabled the commodities to be put on the market at a price especially low.

This contrast, noticeable in the last third of the 19th century, tended to dimmish in the era which opened with the 20th century.

[73] The difference in interest rates between England and the United States became less after 1900, and had less and less effect during the decade preceding the Great War. It ceased entirely during the war itself and the years immediately following. Whatever the future may bring—perhaps equality in interest rates between the two countries, perhaps a slight difference one way or the other—it is tolerably certain that this factor will no longer be of such weight as it was in the earlier period.

A similar contrast, and an illustration of a similar sort, can be found in the effects of railway transportation. The capital account is especially large in railways; the initial investment, the plant, figures to an immensely greater degree than in most industries. The interest charge is therefore an unusually large item in the expenses of production. As regards the labor item, the labor applied to the transportation of an article is as much a part of the total applied to producing it as is the labor of growing or of fashioning (manufacturing). When we envisage the total labor applied to an article produced in England or the United States, we must include the labor in the railway transport of raw materials to the places where they are fashioned and of the marketable goods to the places where they are sold. Now in the United States this labor has been applied with unusual effectiveness in long distance transportation. The United States thus has had a comparative advantage as regards commodities carried over great distances. On the other hand, as regards the interest charge entailed, the United States has been at a disadvantage compared to England. The interest charge, an unusually large item in railway expenses, was long at a higher rate in the United States, and hence had an effect in railway rates, for the same volume of capital, greater than in England. Hence in the United States there was an endeavor to get on with as small a capital-account as possible, and so to lessen the interest burden. But this mitigation of the interest charge, thru a minimizing of capital expenditure, was not at all the most important factor in maintaining for the United States a comparative advantage as regards the item of transportation. The important factor was that of the construction and operation of railways [74] with marked effectiveness of labor; that is, the carrying of many ton miles per unit of labor expended. The great plant was economically laid out and effectively used. The net result was, and probably remains, that in articles which require long inland transportation before they can enter the realm of international trade, the United States had an advantage thruout the period in which the railway has come to be a factor of prime importance; and this notwithstanding the fact that thru much the larger part of the period, i.e. until the close of the 19th century, a higher interest charge on the heavy capital investment was in the nature of a handicap, serving to lessen in some degree the comparative advantage.

Ricardo and his disciples, indeed any economist following the organon of Ricardo, would have criticized sharply the way in which the figures in the earlier part of this chapter are constructed. The basis of the criticism would be that the calculations imply a rise in the price of all goods in consequence of the introduction of interest (i.e. "profits"). The proper treatment is to regard the general level of prices as constant, and to analyze on that basis the connection between prices (values) and the return to capital (interest or "profits"). Then one would have to say that the rates of money wages which were originally set down under the simplest supposition (no capital involved) must be readjusted when capital enters. They must be readjusted downward; wages must be assumed to be lower. If, for example, 30 of copper, made by 10 men, sell for $20, the 10 men can not be getting as much as $20 in wages or $2 per man, since that would leave nothing at all for profit. Wages must be less than $2 a day. If profits are 25 per cent of the wages bill, the rate of wages will be $1.60 and profits will be $0.40; the two items making up the full expense ("cost") of production, equal to the supposed money yield of $20. A mere rise or fall in profits is to be treated as involving a corresponding fall or rise in wages, but not as leading to changes in the prices of goods It is further to be pointed out, following the same analysis, that so far as profits enter to a different extent in one commodity than in another—so far as more capital is used per unit of current labor, or so far as the use of the, same capital is spread over more time,—then a rise or fall in profits would affect the relative prices of goods. A rise or fall thus brought about in any one article would be offset, however, by a corresponding fall or rise in other directions. Changes in the prices of all goods the same way constitute merely a monetary phenomenon, and one not to be confounded with changes in the relative prices of different goods. The proper [75] way to construct illustrative figures such as are used in this chapter is to treat the price level as constant; and the influence of capital should be shown by a process of discounting, as just indicated. So far as a discounting process operates differently on different commodities, there are possibilities of a modifying influence on international trade. And the nature of the modifications will be elicited clearly enough by this method of procedure.

I admit unhesitatingly that the discount method (Ricardo's) is sound. For some purposes, it is the only one that is sound. It is logically the only one tenable for the purposes which Ricardo had in mind when writing the chapter on Value in his Principles: that of analyzing the relations between wages and value, and in general the relations between distribution and value. But it does not seem to me necessarily imposed for the purpose of the present inquiry. It would simply lead to the same results, but by a more troublesome route. Illustrative figures such as I have worked out could be arrived at equally well by the discount method; and they would point to the same general conclusions. They are more easily calculated and more easily followed on the method which I have used, and are equally valid as regards the particular problem in hand; namely, that of showing in what way the item of interest has a modifying effect. True, they make the tacit assumption that advances in the prices of all goods take place as this item is introduced; the advances being greater or less according as the item counts more or less. In order to carry to its logical outcome this assumption of general price advances, it would be necessary to assume also equivalent changes in the monetary supply. For the completeness of the reasoning the reader who is intent on full logical consistency should bear in mind this additional assumption. For the purpose of tracing the effect on international trade—the sole object here in view—it has seemed to me easier and simpler to use the method of supplement, rather than the method of discount. If the object in hand were to consider each and every aspect of a theoretic analysis—monetary theory as well as the theory of distribution, domestic prices as they would be with and without a return to capital—the strict Ricardian procedure would alone be consistent and conclusive.

[76] THE reasoning of the preceding chapters has been based on the assumption that the cost of each and every article is uniform. Changes in the volume of output were supposed to have no effect on the cost per unit. But costs are not necessarily uniform; they are subject to variation according as the total product is large or small. It is incumbent on us to consider the influence on international trade of costs thus varying.

In analyzing this sort of situation I shall return to the highly simplified suppositions made at the start, neglecting the complications and qualifications which have been dealt with in the chapters immediately preceding. That is, labor cost in its simplest form will be considered, uniformity of wages assumed between the laborers (no non-competing groups), capital and a return to capital disregarded. These other factors—which we have seen not to be of such fundamental importance as that of effectiveness of labor—will serve to qualify the conclusions no more and no less than before.

Return once more to the illustrative case already used, showing differences in comparative cost, namely:

The United States has a comparative advantage in the production of wheat; linen will move 'from Germany to the United States, wheat from the United Statee to Germany. The barter terms of trade between the two countries will be 10 of wheat for somewhere between 10 to 15 of linen—11,12, 13,14 linen. The [77] greater the German demand for wheat and the less the American demand for linen, the more favorable to the United States will be the barter terms of trade; the rate between wheat and linen will be nearer to the figure of 15 linen for 10 of wheat. The less the German demand for wheat and the greater the United States demand for linen, the more favorable the terms will be to Germany—the nearer to the figure of 10 of wheat for 10 of linen.

The concrete way in which the conditions of exchange will work out will be thru the range of money incomes in the two countries. A rate advantageous to the United States will be attained by the people of that country thru their having higher money incomes, as well as higher domestic prices. American money incomes will be higher than German in any case; but the difference in favor of the United States will be greater or less according to the play of demand in the two countries for wheat and linen. They will gain as purchasers of import commodities, such as German linen. If on the other hand the conditions of demand should turn favorable to Germany, money incomes and domestic prices will be given an upward trend, and the Germans will secure a larger gain as purchasers of imports, such as American wheat. This is familiar matter; it is restated here by way of introduction to what follows.

Suppose now that hi Germany all wheat is not produced under the same conditions. Suppose that while 10 days produce 10 of wheat on some lands, there are others on which the 10 days produce more. Grade the lands according as the product of this constant amount of labor is 11, 12, 13, 14, 15 of wheat. The price of wheat will be in accord with its cost on the poorest land in use, that on which the 10 days yield but 10 of wheat. The better lands will yield to their owners differential returns, or economic rent.

Under these conditions wheat growing will not cease in Germany after trade with the United States has set in. It would indeed cease if all the wheat were grown under the poorest conditions—were produced at the rate of 10 wheat for 10 days. Germany would then procure her entire supply from the United States in exchange for linen. But as there are varying conditions of supply within her own borders, she would always produce some wheat of her own. The [78] very best German lands, those on which 10 days of labor yielded 15 of wheat, could always hold their own against American competition. Those on which the yield was less than 15 might or might not succumb under American competition. The continued cultivation of wheat on them would depend on the terms of exchange between the two countries. If the terms were 14 of wheat for 10 of linen (favorable to Germany), only those German lands which produced as much as 14 could continue in face of American competition ; those on which the yield was less than 14 (13,12,11) would find themselves forced out. If, on the other hand, the terms of exchange were not favorable to Germany—if she got for her 10 of linen only 11 of wheat—a poorer grade of German land could continue the production of wheat. The more wheat the United States gives in exchange for linen, the more will Germany restrict her production of wheat to those lands on which her labor is least ineffective. That is, where it is least ineffective compared with American labor in wheat-raising; most effective, compared with other German labor in wheat. Barter terms of trade which are favorable to Germany will mean that the conditions are not favorable for the maintenance of her wheat production.

Obviously this means also that the rent of wheat lands in Germany depends on these same terms of trade. If the terms are unfavorable to Germany,—if she gets but 11 of American wheat in exchange for her 10 of linen—the wheat-growers on her poorer lands will remain in the market, and the differential advantage of the better lands will be little impaired. Their rent will be the excess of product over 11 of wheat, instead of the excess over 10; the margin of cultivation will move only from 10 to 11. But if the terms are favorable to Germany,—if she gets 14 of wheat for her 10 of linen—more of her wheat lands will be forced out of cultivation, and those which continue to grow wheat will afford less rent. The grade on which 14 wheat are produced for 10 days of labor will afford no rent at all; that on which 15 wheat are produced will afford a rent of only 1.

Under a regime of prices all these results will work themselves out as they would under a barter regime. If the terms of trade are [79] favorable to the United States,—if she gets 14 of linen for 10 of wheat—money incomes are comparatively high in the United States, comparatively low in Germany. Wheat then is comparatively high in price in the United States, and is at the same comparatively high price in Germany. Linen is at a comparatively low price in Germany, and at the same low price in the United States. As consumers of linen, the Americans gain from their high money incomes; as consumers of wheat, the Germans lose from their low money incomes. But as producers of wheat, the American wheat-growers are under a handicap in selling their wheat in Germany, They cannot sell so much, nor can they displace as many German wheat-growers as they could if their money incomes and their wheat prices were lower. And the Germans as producers of wheat are not so hard pressed by American competition as they would be if their (the German) money incomes were higher. The low rates of money wages lessen their expenses of production, and wheat lands which would go out if money wages were higher are able to hold their own and maintain themselves in face of American competition.

In the talk of the man on the street, and also, unfortunately, in the reasonings of pretentious books on economics, consequences of this kind are dealt with as if they indicated a disadvantage to the United States and an advantage to Germany. The American wheat-growers find in higher money wages an obstacle to the cheap production of wheat and to the extension of exports; this is bad for the United States. The German wheat growers find in lower money wages an aid in meeting foreign competition; this is good for Germany. The man on the street almost invariably has the mercantilist point of view: exports are to be promoted, but domestic production is also to be safeguarded against competing imports. Not a few economists share these prepossessions, sometimes deliberately, more often thru a lack of sustained and consistent thinking. True, no economist, and indeed no thinking person, would deny that high money wages, combined with low prices of goods, bring material prosperity; but, when faced by a concrete situation, few accede readily to the conclusion that high [80] money wages, even tho they compel some domestic producers to lessen or abandon their output, are still the result and the indication of better conditions of the community at large.

Another problem, and one on which there is much more occasion for difference of opinion among the discerning, concerns the range of industries in which we may expect to find varying costs. Wheat has been selected for illustration, because it is a commonplace in economics that agricultural commodities are usually produced at varying costs. But are not other classes of articles also produced at varying costs? The more ample information yielded in recent times by statistical inquiry, and our greater familiarity with the actual conditions of industry, both point to the conclusion that in manufacturing industries also we find, at any given tune, costs not uniform. Side by side there are effective and ineffective producers. Must we not assume for the entire range of industry conditions like" those assumed for wheat? And is not the theory of international trade to be readjusted accordingly?

It would carry us far from the main topics of the present volume to consider this question in all its ramifications. The discussion involves moot points, and illustrates once more the impossibility of separating the theory of international trade from the general problems of economics. I content myself with a summary statement of the grounds for a negative answer; negative, that is, as regards any considerable modification of the theory of international trade.

Manufacturing industries do show varying costs. Uniform costs are never found. But the causes of variation are different from those found in the extractive industries; the persistence of the phenomenon is due to different causes; and the consequences are different, both as regards domestic trade and international trade.

The main cause of variation is in the personal element. Some managers of industry are more efficient than others, and cost of production at their hands is less. The explanations which usually figure in the discussions on this topic—better location of the low-cost establishments, better access to materials, better plant, better organization—are reducible to this one dominant element, the [81] differences in managerial capacity. The reason why some establishments, for example, are better located than others is at bottom the same as that why some are better organized than others: the managers are more shrewd and capable. And these more capable managers get a differential return analogous to rent. It is analogous to economic rent, that is, in industries where there is nothing which could in strictness be called a monopoly—where there is no control of supply in any single hand, but a free field for all who care to enter.

The phenomenon differs from that of economic rent, however, in that these powers of superior productiveness are transferable. The abler business man is not bound to any industry or any field. He roams at large, turning his faculties in whatever direction he finds they tell most profitably. The case is otherwise with land and natural agents. If a landowner finds his differential gain lessened, he must accept the situation once for all and submit to the loss. To some extent he can indeed turn the land to one crop or another; but the possibilities of such shifts are limited, and they mean, not that loss is avoided, but only that it is perhaps made smaller. The landowner cannot transfer the superior powers of his acres to other acres or to a manufacturing industry. The superior business manager, however, if he finds that his powers are exercised with less effect in one industry than another, turns from the less profitable to the more profitable.

It is true that a superior business man may secure ordinary profits (i.e. non-superior profits) in an industry which would yield no profits at all to the non-superior business man. Tho he engage in an industry which possesses no comparative advantage of the sort that has been illustrated and discussed in these pages, he may succeed, notwithstanding the absence of advantage, in making both ends meet and even in clearing a profit for himself. But he cannot clear as much for himself as he could in industries adapted to the general industrial advantages of the country. And the superior business man will not ordinarily turn to the ill-adapted industries. One of the signs of superiority is the very fact that he has an eye for the more promising possibilities. He sees better than most what is [82] likely to be suited to profitable operation and what is not. It is true that accident, inheritance, early misjudgment, may start him in an industry which fails to give full scope to his abilities; and there are plenty of cases in which capable men, once started the wrong way, remain in the industry of first choice. But in the main it is otherwise. The business man of higher grade turns to operations that afford full scope to his abilities; and the more outstanding are those abilities, the less likely are they to be misapplied to unfruitful industries.

An excellent illustration of the consequences of this transferability, and of the general theorem here advanced, is found in the working of protective duties. Suppose an industry to be fostered by protection; for example, the linen industry in the United States operating under conditions of comparative disadvantage such as are shown in our figures. Suppose then that the protective duties are abolished; will the linen industry disappear from the country? It is entirely possible that an examination of the expenses of production of the several linen-making establishments will show them to be producing at varying money costs. Some among them will be found to be producing so cheaply that they can hold their own, and continue to make ordinary profits, even under free trade. But these will be the superior establishments, managed with superior ability; and their managers, even tho they meet their expenses of production and eke out a profit, will be getting a return less than in proportion to their powers. Sooner or later they will shift. The transfer to other industries will not take place easily or quickly, but in the end it will come. There may be in such a protected industry older men who, while capable, yet are too advanced in years for a complete change of base; and these will very probably remain where they are. But younger men of the same stamp will not be drawn to it; the industry will languish and in time will disappear.

Still another aspect of manufacturing industries must be considered. They are often said to present the conditions not of diminishing but of increasing returns. Is not our reasoning in [83] need of elaboration and modification on this score? Just as there are special conclusions for the industries in which cost increases as output enlarges, so we might expect other special conclusions for commodities in which cost declines with enlarged output.

Two things must here be noted: first, what exactly is meant by a law or tendency to increasing returns; and second, what is the effect of this sort of tendency in international trade.

On the first topic some distinctions familiar in economic theory must be recalled. Any "law" of increasing returns means that all costs go down. The law is not at all the converse of that of diminishing returns in agriculture. In the latter, an increase of the output from a given plot or given area of land entails as a necessary corollary that, while the additional supplies are got at higher cost, the previous supplies continue to be got at cost unchanged. Therefore there are varying costs—some costs persistently higher than others. In the apparently opposite case of increasing returns, there are no persisting differences. True there is lower cost with enlargement of output; but it is the entire supply which is produced at the lower cost. True, not all will be produced at lower cost immediately; but in the end it will. As has just been explained, there will probably be a transition period of varying costs. An improvement which lessens costs is almost invariably introduced gradually, first in one establishment, then in another. For a while costs will be lower in the forward than in the lagging industry. The ultimate effect will be a decline all around.

To come now to the main general conclusion which bears on the problems of international trade. Such a decline, when it has permeated the whole of an industry, may mean a change in its costs relatively to other industries. It may mean a new alignment of comparative costs, and accordingly may alter the conditions under which international trade is carried on. Such consequences, however, are not of a novel kind, and call for no new analysis. With the irregular progress of the arts, the conditions of comparative advantage are subject to constant modification; but these changes, while they lead to new conditions, involve merely the application of familiar reasoning to the changed situation.

[84] Still a further remark, however, is to be made, one which illustrates once again the connection between the various parts of the structure of economic theory. Professor Marshall has taught us to distinguish between "internal" and "external" economies—between those improvements and lowered costs which arise primarily within each several establishment or industry and those which are the outcome chiefly of forces outside the establishment or industry. Concerning internal economies, such as mechanical inventions, scientific discoveries, better organization within the plant, little in the way of a trend or law can be made out. They come or do not come, as it so happens. But external economies have a trend which is predictable. They are in themselves the result of larger aggregate output. The mere fact that there is a larger total product of plows, motor cars, safety razors, tends to make each unit cheaper. Greater specialization and subdivision of labor become possible; there is a greater pervasive facility of industrial advance.

This is not the place for considering disputed matters relating to the general tendency. What concerns us is that it has some special consequences for the international trade of those countries which export manufactured articles. In agriculture, external economies are not indeed lacking in effect; larger aggregate output does bring into action some causes of decrease in cost (better roads, for example, or cheaper plows); but these are offset, in part or completely, by the tendency to diminishing returns in those operations which have to do with the direct culture of the soil. It is not easy to say under what conditions external economies may be so effective that agricultural costs on added yields from the same soil tend to increase or decrease as aggregate output enlarges. Something of this kind may happen for a while in a country which has intelligent and progressive population. But it would seem that with growth of numbers agricultural costs must increase relatively; that is, tho they fall, they will not fall as much as will the costs of manufactured articles. A country which is growing fast; whose industries are largely manufactures; whose exports of such goods are large; whose total output of them is increasing; whose costs [85] per unit are therefore going down—such a country will not only have a comparative advantage in manufactured goods, but will probably have a growing comparative advantage. The more it produces of such goods, the greater may be its advantage for exporting them; and hence it will turn its labor cumulatively in this direction. While it will have costs which (in the long run) are uniform for each several article of export, its costs will tend also to decline for each several article. In that sense—considering successive stages, not any given stage—it will have varying costs.

This sort of advantage, even tho it generates itself and goes on crescendo, does not persist indefinitely. It rests primarily on human causes, not on those of the physical world without. It is subject to the vicissitudes of industry and in some degree to man's deliberate action. England seems to have had some cumulative advantage of this kind during the first half or two-thirds of the 19th century. As time went on, other countries entered on the same paths; and they were probably aided in doing so by protective duties on their manufactures, that is, by deliberate action. At all events the international division of labor, while still affected by England's matured position, was gradually controlled more and more by forces of a deeper and more permanent character, and this particular sort of advantage no longer played a part in shaping England's foreign trade. At a later period, during the closing years of the 19th century and the opening years of the 20th, the United States'also experienced a burst of industrial advance, and with it an astonishing development of external economies; and with this again a re-alignment of the effectiveness of labor in the several branches of production. Here, too, while agriculture was affected somewhat, manufactures were affected more. The proportion of manufactured exports tended to increase; and what was no less significant, the proportion of manufactures among the imports tended to decrease. Here, too, the change, cumulative tho its moving forces were, was not likely to progress indefinitely. As in the case of England, it did leave its permanent impress on the international trade of the country, as well as on its domestic trade. But as time went on, other countries were likely to enter on similar [86] paths; and then once more it became a question what were the long sustained currents in the movement of goods from country to country.

Mining industries present a peculiar case,⁰¹⁴ in international trade as well as in domestic. It is not easy to say whether they are to be regarded as analogous to agriculture or to manufactures. They are extractive industries, and therein like agriculture; but they are not subject to any "law" of diminishing return; at all events, to none of the same kind that bears on agriculture. It is true that in mining we find at any given period of time varying costs quite as sharply as in agriculture, and more sharply than in manufactures. But mining operations are not subject to the kind of pressure which appears when men cultivate the soil—when they use land as a means for yielding crops perennially. In mining there is a fixed store, not an instrument for transmuting matter. And the mines which are known and are in use at any one time, while they vary in richness, show no certain trend toward greater or less richness. All that we are sure of is that each and every single mine will sooner or later be exhausted; but it may continue to be unvarying in richness until it gives out. Still more uncertain is it whether new mines will be discovered, and whether they will prove worse or better than those previously worked. We have further to consider that while the variations between the costs at the several sources of supply are due to physical causes such as differences in richness and accessibility, their availability depends markedly on the same human factors which are outstanding in manufactures, the ability and venturesomeness of business leaders. All in all, it is doubtful whether we can speak in strictness of any tendency to constant returns or diminishing returns or increasing returns in mining. For the theory of international trade we must be satisfied with some general empirical results. The tangled forces that act in the mineral industries—the mysteries of the earth's crust, unpredictable new finds, the progress of science, the progress of business management and industrial organization—bring it about that at [87] one time this country, at another time that one, has special effectiveness, absolute or comparative, for copper, silver, tin, coal, iron. A de facto situation appears in the matter of comparative costs and comparative advantages, and international trade is shaped accordingly. Whether the trade is likely to continue for a given country on the same lines in the future as in the past or present, we have no means even of guessing. In the case of agricultural products we may expect, on grounds of general reasoning, that a change in demand or in supply will lead to certain permanent alterations in the positions and advantages of the several trading countries. But we can hardly apply such theorizing to the products of mines. We have simply to accept the situation as it happens to develop at the given time and place.

[88] IN the preceding chapters it has been assumed that there were but two commodities in the trade between the countries and one commodity exported from each. Any given country, however, exports not one article, but a number. This circumstance in itself would not necessarily point to modifications of the reasoning. If the several articles were all produced under the same conditions of advantage or disadvantage, they could be treated as one. But it is not to be assumed that all are alike in this regard—that there is in each and every industry of a given country the same trend in the effectiveness of its labor compared to effectiveness in other countries. It is almost certain that a country will have a greater superiority in some directions than in others. Once more we are compelled to modify our reasoning and amplify our deductions by introducing supplementary hypotheses; reshaping the conclusions reached on the simplest assumptions by introducing further assumptions such as to bring us closer to the realities.

Suppose that there are not two commodities but three; and suppose further that we find not the same relations of comparative effectiveness between the trading countries, but a graded situation. Let the three commodities be wheat, woolen cloth, and linen. The following figures will serve for illustration and for analysis.

[89] Observe that the effectiveness of labor in the United States is greater thruout than its effectiveness in Germany. The same amount of labor produces in the United States more of wheat than in Germany, more of linen, more of cloth. But the superiority is greatest in wheat; it is less in linen; it is least in cloth. What sort of trade will emerge ?

Consider first the possibilities of barter; and begin by considering these possibilities separately for the three pairs of commodities; namely, wheat as against cloth, wheat as against linen, and cloth as against linen.

(1) Suppose the situation were wheat against cloth.

The United States and Germany would both gain at any barter terms of trade intermediate between 10 and 18 of German cloth for 10 of American wheat; the gain to the United States being greatest if nearly 18 of cloth were got for 10 of wheat, and that to Germany greatest if but little more than 10 of cloth were given for 10 of wheat.

(2) Suppose the situation were simply wheat against linen.

The United States and Germany would both gain at any barter terms of trade intermediate between 10 and 15 of German linen for 10 of American wheat. The nearer the terms were to 15 linen for 10 of wheat, the more the United States would gain; the nearer to 10 of linen for 10 of wheat, the more Germany would gain.

(3) Suppose finally that the situation were cloth against linen.

[90] The United States and Germany would both gain if linen went from the United States to Germany. At any terms of trade between 10 and 12 of German cloth for 10 of American linen, both countries would gain.

Observe that in case (2) linen moves from Germany to the United States, to the gain of both countries. In case (3) linen moves the other way, from the United States to Germany, also to the gain of both countries. What determines whether one or the other sort of movement takes place?

The answer is: the outcome depends on the play of demand. The nature of the trade, and especially the movement of linen, will depend on the demand of the United States for her own wheat as compared with her demand for the two commodities which may conceivably be secured from Germany, cloth and linen; and on the demand of Germany for her own cloth as compared with her demand for the two commodities which may conceivably be secured from the United States. The barter terms of trade may be so favorable to the United States that she will find it advantageous to procure from Germany both linen and cloth. Or the terms may be so little favorable to the United States, and her margin of gain so narrow that, tho it remains advantageous to get from Germany, that commodity (cloth) which Germany can supply most cheaply, it will not be advantageous to get from Germany the other commodity (linen) which Germany can supply less cheaply. It is in cloth that Germany has her greatest effectiveness of labor; it is here that she has the least of her inferior disadvantages; and cloth will move from Germany to the United States even tho the barter terms of trade are not favorable to the United States. Nay, as will be presently shown, those terms may be so unfavorable to the United States that she will gain by actually exporting linen to Germany. Linen may move either way, according as the barter terms of trade vary. There are thus three possible cases:

I. Both cloth and linen may move from Germany to the United States. As regards the exchange between German cloth and American wheat, we have seen that both countries would gain if [91] American wheat were exchanged for German cloth at any figure between 18 and 10 of cloth against 10 of wheat. Suppose the barter terms of trade to be advantageous to the United States; that by sending 10 wheat to Germany she gets 15 of cloth. In Germany 15 of cloth are produced with the same labor as 12-1/2 of linen (18:15::15:12-1/2), and it is immaterial to Germany whether she gives 15 of cloth or 12-1/2 of linen for the 10 of wheat. The United States gains by either act of exchange: if she gets 15 of cloth for 10 of wheat, she gains 5 of cloth; and if she gets 12-1/2 of linen for 10 of wheat, she gains 2-1/2 of linen. That particular combination or proportioning of the commodities (15 of German cloth and 12-1/2 of German linen for every 20 of American wheat) may precisely suit the mutual tastes or demands. Both countries will then gain if not cloth only, but linen also, moves from Germany to the United States.

II. Take now a situation toward the other extreme, one in which the barter terms of trade are favorable not to the United States but to Germany. Suppose the United States gets in exchange for 10 of wheat no more than 11 of cloth; the United States thus gaining from the operation only 1 of cloth. In Germany 11 of cloth are produced with the same labor as 9-1/6 of linen (18:15::11:9-1/6), and the American wheat, which exchanges for 11 of cloth, would exchange at German rates for only 9-1/6 of linen. Obviously the Americans get more linen (10) for their 10 days of labor by producing it directly than by procuring it from Germany. But more. The United States now not only will find it worth while to produce her own linen; she will gain by exporting it to Germany and taking cloth in exchange. It is immaterial to the United States whether she sends 10 of wheat or 10 of linen to Germany—both are produced with the same labor. Within Germany, however, 12 of cloth exchange for 10 of linen, and therefore 11 of cloth exchange for 9-1/6 linen. The United States by sending 9-1/6 linen to Germany can get 11 of cloth in exchange. This particular combination or proportioning of commodities (10 of American wheat together with 9-1/6 of American linen in exchange for every 22 of German cloth) may again precisely suit the conditions of mutual [92] demand; both wheat and linen then move from the United States to Germany.

III. Lastly, take an intermediate case—the intermediate case. Suppose the barter terms of trade to be 10 of wheat for 12 of cloth (less than 15 of cloth as in our first case, and more than 11 of cloth as in the second). Within Germany 12 of cloth are produced with the same labor as 10 of linen. The United States, sending 10 of wheat to Germany, and getting 12 of cloth in exchange, might indeed get also 10 of German linen in exchange. But 10 of linen are produced in the United States with the same labor as 10 of wheat; there is no gain to the United States. Germany might send 12 of cloth to the United States and would then receive in exchange 10 of linen. But 10 of linen and 12 of cloth are both produced in Germany with the same amount of labor (6-2/3 days); and there would be no gain to Germany. Neither country would find it worth while to send linen to the other. The only gainful exchange is that of German cloth for American wheat. The case is, in a sense, the mid-way or balancing one, that in which one commodity (linen) remains where it is, while the others move to and fro.

This train of consequences from the play of varying demand is dependent on the toted demand in each country for the products of the other. To speak more accurately, it is dependent on the state of demand among the inhabitants of the two countries for each and every one of the commodities which they might exchange. The Americans may care for cloth and linen so much, and for wheat so little, that they will offer wheat for the other two commodities on terms that make it worth while for the Germans to send both cloth and linen to the United States. Or, at the other extreme, the Americans may care for cloth and linen so little, and for their own wheat so much, that they will indeed take cloth in exchange for some wheat, but will take no linen; nay, may prefer to send some linen of their own make in exchange for cloth.

These possible relations may now be expressed in terms of prices and money incomes. Let it be remembered, in considering the [93] figures which follow, that a relatively high rate of money wages in the United States—a considerable gap between American and German wages—signifies that the United States secures the larger share of the possible gain from the trade; whereas relatively lower rates in the United States—a smaller gap in money wages—signify that the United States secures the smaller share. For simplicity, the figures will be arranged on the basis of keeping money wages in the United States at a constant figure, namely $2.00 a day^v. The changes which serve to illustrate the different possibilities are here confined to Germany, where money wages become lower as the terms are less favorable to her, higher as they become more favorable.

Again we take the three possible cases.

(1) Suppose first a wide gap between German and American money wages. Let wages in the United States be $2.00, wages in Germany $1.20. We have then:

Wheat is produced at lower money cost in the United States than in Germany and moves from the United States to Germany. Both linen and cloth are produced more cheaply in Germany and move thence to the United States. The United States, while gaining thru the importation of both, evidently gains more from the importation of cloth than from that of linen. She gets her cloth from Germany for $0.66-2/3 whereas the price at which cloth can be made in the United States is $1.00. She gets her linen from Germany for $0.80; less than the American supply price of $1.00, but not as much below that price as in the case of cloth. Germany gains by a cheapening of wheat to the amount of $0.20. It would cost her $1.20 to produce wheat at home; she procures it from the United States at the price of $1.00.

[94] Consider now the barter terms of trade which obtain under these circumstances. In both countries wheat sells for $1.00 and cloth for $0.66f, these being the "world prices." In terms of commodities, 10 of wheat exchange for 15 of cloth. The five days' labor which in the United States produce 10 of wheat would yield, if applied to cloth, 10 of cloth also; the United States, getting 15 of cloth for her 10 of wheat, gains 5 of cloth. Germany, on the other hand, gains 3 of cloth. With 10 days' labor she could produce 18 of cloth or 10 of wheat; with only 15 of cloth she gets 10 of wheat.

Linen sells for $0.80 in both countries; with wheat at $1.00 the terms of trade are 12? of linen for 10 of wheat. The United States gets 12^- of linen for 10 of wheat, the product of 5 days' labor, whereas that labor would produce at home only 10 of linen. The gain is 2? of linen for the United States. And Germany gains the same: the difference between 15 of linen and 12?.

(2) Next, assume barter terms of trade between the two countries which are more favorable to Germany; the evidence of the more favorable terms being higher money wages in that country. Suppose German wages to be not at a figure somewhat low ($1.20), but much higher, say $1.60. Wages in the United States we assume to remain at $2.00. Then we have:

Both wheat and linen are now cheaper in the United States than in Germany. The difference, of course, is greater for wheat, whose domestic supply price in Germany is $1.60, whereas it can be got from the United States for $1.00. Linen can be produced in Germany at a money cost of $1.06f; but it is obtainable from the United States at a slightly lower figure—$1.00. Cloth still moves [95] from Germany to the United States, being put on the market in Germany at the price of $0.89 as against an American money cost of $1.00. The marked change from the previous situation is that linen, which before moved from Germany to the United States, now moves from the United States to Germany.

The barter terms of trade are, as between wheat and cloth, 10 of American wheat for 11.2 of German cloth; as between linen and cloth, 10 of American linen for 11.2 ef German cloth. The United States thus gets for 10 of wheat only 11.2 of cloth, as against 15 of cloth under the previous conditions. Under those conditions she had such favorable terms for cloth as to lead her to confine her labor to wheat alone. Linen was then got more cheaply from Germany than by domestic production. Now the case for linen is reversed. It can no longer be got at lower price from Germany. On the contrary, it can be produced at so much lower money cost at home that it is actually exported to Germany. The explanation for this overturn, to repeat, is that the United States no longer gets the lion's share of the potential gain divisible between the two countries. She did get that preponderance of gain when her wheat was greatly in demand in Germany, and when on her part she did not readily take either cloth or linen in exchange. The relative states of demand have changed; the United States wants more cloth, Germany wants less wheat; to get what she wants, the United States finds it advantageous to send not only wheat, but linen also. The money wages and money prices constitute the mechanism by which those new conditions are transformed into actualities; but the fundamental cause of the change is the altered state of demand for the several articles in the two countries.

(3) Finally, the intermediate case. Let German wages be $1.50 a day—not so high as $1:60, not so low as $1.20. Then we have:

[96] The price situation is simple. Wheat is produced more cheaply in the United States than in Germany—$1.00 in the United States and $1.50 in Germany. Cloth is produced more cheaply in Germany—$0.83-1/3 there as compared to $1.00 in the United States. Linen, however, has the same money cost of production in both countries; namely, $1.00. Wheat moves from the United States to Germany. Cloth moves from Germany to the United States. Linen moves neither way; each country produces for itself the linen that it consumes.

The barter terms of trade into which these prices ($1.00 for American wheat, $0.83-1/3 for German cloth) resolve themselves, are 10 of wheat for 12 of cloth. On these terms, it is a matter of indifference to Germany whether for the 10 of American wheat she exchanges 10 of her linen or 12 of her cloth since in Germany 10 linen is equal to 12 cloth. For the United States, however, it is more advantageous to take the 12 German cloth in exchange for her own wheat inasmuch as in the United States 10 linen is equal to only 10 of cloth. In this case, therefore, the American situation would be decisive, and American wheat would flow to Germany and German cloth to the United States.⁰¹⁶

[97] STILL another modifying circumstance is now to be introduced. So far the problems have been treated as if there were but two countries. We proceed to consider some changes or qualifications which appear when we have not a single country exchanging with one other, but several countries competing with each other in supplying another country. Here, as in the last chapter, the procedure will be that of considering labor costs alone, and these in their simplest aspects, the reader being assumed to bear in mind that other factors (such as non-competing groups among laborers, capital and the return on it, varying costs) complicate the situation and may modify the results. The analysis of the fundamental factor of labor costs, taken by itself, serves to bring out the essentials for the problem here in hand.

First, suppose a case in which there are two countries on the one side, a single country on the other. Let the two be the United States and Russia; the single one, England. Let the conditions be such that both the United States and Russia have a comparative advantage over England in wheat, England a comparative advantage over them in cloth. In figures, for example, thus:

A glance shows that both the United States and Russia can trade to advantage with England, exporting wheat and getting cloth in exchange. Both have a comparative advantage over England in wheat, tho not of precisely the same kind. The United States has [98] a superior advantage in wheat; Russia has an inferior disadvantage. Tho the United States produces both wheat and cloth with less labor than England, the effectiveness of her labor is particularly great in wheat. With the same labor, the output of wheat is twice as great as in England (20 to 10), while that of cloth is only one-third greater (20 to 15). Russia has no superiority over England in either commodity; but she has equal effectiveness in wheat (10 to 10), with a less effectiveness in cloth (10 to 15).

Russia and the United States, it is obvious, have no occasion to trade with each other. They present the simple case, already considered sufficiently, of equal differences in costs. The effectiveness of labor is twice as great all around in the United States as in Russia. Neither country would find it worth while to exchange with the other. The United States is the more prosperous, Russia the less prosperous. Were they alone, and England out of the case, neither would pay attention to the other; neither would be better off or worse off because of the presence of the other.

The terms of trade possible under these conditions would be 10 of wheat for anywhere between 11 and 14 of cloth. These terms, that is, would be possible in trade between the United States and England, and also in trade between Russia and England. England would exchange with each of the others on the same terms. So far as concerns the gain ascribable to international trade, both the United States and Russia would be on a footing of precise equality: their income in terms of the cloth secured from England would be enlarged to precisely the same extent over and above what that income would have been without the trade.

Express the same situation in prices and money incomes. As we have already seen, the double effectiveness of American labor as compared with Russian would cause money wages to be twice as high in the United States as in Russia; while the relations between the effectiveness of labor in England and in the other two countries would bring it about that money wages would be higher in England than in Russia, lower than in the United States. We may have, for example:

[99]

The supply price of wheat is the same in the United States and Russia ($1.00), and wheat will sell at that price not only in these countries, but in England also. Russia cannot undersell the United States in wheat, even tho her wages are but half of American wages; since the effectiveness of her labor is also one-half. Cloth is produced at a cheaper price in England than in the other two countries; and English cloth will be exported to both, and will be sold in both at the same price—$0.83-1/3. The American purchasers, tho they pay for the English cloth the same price as the Russians, have money incomes twice as large, and therefore are better off as purchasers. Their better situation, however, is obviously due to the same cause as the generally larger prosperity of the United States; it is the result of the greater effectiveness of labor in wheat. So far as concerns the terms on which the United States gets her cloth from England, she is on precisely the same footing as Russia.

Construct now an international balance of payments based on these price relations. Suppose that:

Russia and the United States (between them) buy from England
15 million cloth at $0.83-1/3 = $12,500,000

Russia and the United States (between them) sell to England
12-1/2 million wheat at $1.00 = $12,500,000

The two money totals are the same. An equilibrium of payments is established, foreign exchange is at par, no specie moves, the wheat and the cloth pay for each other. Wheat to the amount of 12? million bushels is exchanged for cloth to the amount of 15 million yards. That is,

12-1/2 wheat = 15 cloth, or 10 wheat = 12 cloth

Of the possible terms of trade (10 wheat for anything more than [100] 10 or less than 15 cloth) that at which the countries are bartering in fact is 12.

Suppose now a change in the conditions of demand. Assume that at the price of $0.83-1/3 for cloth, more cloth than the 15 million yards can be sold in the United States and Russia. The increase in quantity demanded at that price may come from Russia alone, or from the United States alone, or partly from each of them. Whatever the region whence the increased demand appears, the result is that England sells more cloth. Her exports then exceed her imports in money value, and specie flows to her from the other countries. The consequences are familiar; prices and money wages rise in England, fall in the countries with which she is trading, and changes of this kind go on until a new equilibrium is established. The new states of wages and prices may be exemplified thus:

Money wages have fallen both in the United States and Russia; from $2.00 to $1.90 in the United States, from $1.00 to $0.95 in Russia. As purchasers of cloth, both Russians and Americans are worse off than before; their money incomes are lower, the price of cloth is higher. Money wages in England on the other hand have risen, and the English are better off as purchasers of wheat.

The readjusted equilibrium of international payments may then be exemplified thus:

Russia and the United States (between them) buy from England
20 million cloth at $0.90 = $18,000,000

Russia and the United States (between them) sell to England
19 million wheat at $0.95 = $18,000,000⁰¹⁷

It now appears that 19 million bushels of wheat are exchanged by Russia and the United States for 20 million yards of English cloth.

[101] The barter terms of trade, that is, become 19 wheat for 20 cloth, or 10 wheat for 10-10/19 cloth—very nearly 10 for 10-1/2. Before the increase in demand for cloth set in, the barter terms of trade had been 10 wheat for 12 cloth; they are now 10 for 10-1/2. The English get the same quantity of wheat for a less quantity of cloth; they get a larger share than before of the possible gain from the trade.

It matters not, to repeat, whether the change in demand takes place solely in the United States, solely in Russia, or partly in one and partly in the other. An increase of Russian demand operates to make the terms less favorable to the Americans, even tho in the United States alone nothing has happened that would change the situation.

This sort of case, in the two possible phases here worked out, serves to illustrate a general proposition which played its part in the exposition of the classic doctrine. "There are two senses in which a country obtains its commodities cheaper by foreign trade; in the sense of value and in the sense of cost."⁰¹⁸ In the sense of cost, the United States thruout is getting its linen cheaper than Russia. In the sense of value, both the United States and Russia get the English linen on the same terms. Cheapness in the sense of costs depends on the amount of labor given to the exported commodities; this is less in the United States than in Russia. Cheapness in the sense of value depends on the barter terms of trade between the exports and the imports; at any one time this is the same for the United States and for Russia, but varies at different times according to the conditions of demand for wheat on the one hand, for linen on the other.

Obviously it is more probable that the barter terms of trade will be favorable to England and that she will get her imports (wheat) cheaper in the sense of value, if there be not one country buying cloth from her but two or more. The greater the number of purchasers of her cloth, the larger the quantity that will be taken at a given price. If the other two countries, United States and Russia, have the same population and are alike in the demand of their peoples (in their demand schedules) for cloth, then the two of them

[102] will take twice as much of cloth at the same price as either of them would take alone; and the barter terms of trade will become more favorable to England.

Turn now to a case of a different kind, and one more complicated. It is exemplified by the following figures:

Here each of the countries has a situation as regards the relative costs of the two articles which is different from that of either of the others. The United States has a comparative advantage over both England and Germany in wheat, and might exchange wheat for cloth with either or with both. England has a comparative advantage in cloth not only as against the United States but as against Germany also; and she might send cloth to either in exchange for wheat. Germany might send cloth to the United States in exchange for wheat; but she might also send wheat to England in exchange for cloth. The limits within which the barter terms of trade thruout the trading area could establish themselves are 10 cloth for 10 wheat at the lowest, 15 cloth for 10 wheat at the highest. At any rate between these limits (i.e. wheat exchanging at the rates of 11, 12, 13, 14 for cloth) there will be trade. But which of these several possibilities will emerge?

The situation is essentially the same as that considered in the first part of this chapter; indeed, is no more than a variant. The answer again is that the outcome depends on the state of demand between the countries. As before, there are three possible cases.

(1) Suppose the barter terms of trade to be unfavorable to the United States—such as would exist if the demand of the United States for cloth were great, the demand of England and Germany for wheat small. Suppose it to be 10 wheat for 11 cloth. Both Germany and England would then send cloth to the United States, [103] and the United States would send wheat to both in exchange. England would gain more from the operation than Germany; she would gain the difference between 15 and 11. Ten days' labor in England yields 10 of wheat and 15 of cloth; if England gets 10 wheat for 11 of cloth, she gains the difference between 15 and 11. Ten days' labor in Germany yields 10 of wheat and 13 of cloth; if Germany gets 10 of wheat for 11 of cloth, she gains the difference between 13 and 11. The United States would gain the difference between 10 and 11.

(2) Next suppose that the terms of trade become distinctly favorable to the United States,—that she gets for 10 wheat as much as 14 cloth. Then England would send cloth to the United States and the United States wheat to England. But at these terms England and Germany would also exchange. Germany would gain by sending 10 wheat to England and getting 14 cloth in exchange. Since the given labor (10 days) would produce in Germany only 13 cloth, England would gain similarly; with 10 days' labor she could produce at home 10 wheat or 15 cloth; if she gets 10 wheat for less than 15 cloth, she gives her labor more advantageously to producing cloth only. At the rate of 14 cloth for 10 wheat,'then, both the United States and Germany would send wheat to England, and England would send cloth to both; nor would any cloth be made either in Germany or the United States. No trade would take place between the United States and Germany, notwithstanding the fact that trade between them would develop if England were out of the way and they were confronted merely with each other.

(3) Suppose now the intermediate stage, that at which the terms of trade are exactly 10 wheat for 13 linen. England then will send cloth to the United States and the United States will send wheat in exchange. But for Germany the situation would be one of indifference. If she were to send 13 cloth to the United States she would secure (at the rate established between England and the United States) 10 of wheat, or precisely the same amount of wheat as she could produce at home with the labor given to producing the 13 of cloth. The trade would be between the United States and [104] England only. Any deviation from these terms (13 cloth for 10 wheat) would cause Germany to enter. If the cloth given in exchange for 10 wheat were more than 13 cloth, Germany would turn from cloth to wheat and would send wheat to England. If it were less than 13 cloth, Germany would turn from wheat to cloth, and would send cloth to the United States. At the precise figure of 13 cloth she would have no inducement for concerning herself with the other countries at all, and would go her way, producing for herself both cloth and wheat. The only trade would be between the United States and England.

These suppositions, like the various others which have been considered in the preceding pages, can be put in terms of money prices and money incomes. At the risk of wearying the reader, I will indicate how money wages and money prices might shape themselves in the three countries in the several cases just described.

(1) Suppose that money wages and domestic supply prices in the countries are as follows:

The domestic supply price of cloth is the same in England and in Germany—$0.93. It is lower than the domestic supply price of cloth in the United States; and both German and English cloth will be sold in the United States at a price which no American cloth maker could meet. The Americans would get their cloth for $0.93 by importation, instead of paying $1.00 for it, as they would if it were made at home. Both Germany and England would get American wheat for $1.00. Wheat, if grown in England, would entail a money cost of $1.40; if grown in Germany, would entail a money cost of $1.21. In other words, England would gain the difference between $1.40 and $1.00, and Germany the difference between $1.21 and $1.00. Both gain, but England gains more.

[105] The barter terms of trade would obviously be 10 wheat = 10.7+ of cloth; this being the ratio in terms of physical units of the price relations—$0.93 for cloth and $1.00 for wheat. That is, 10 wheat exchange for less than 13 cloth. Trade on this basis, as we have seen, gives a large share of the possible gain to England and Germany ; a comparatively small one to the United States, even tho one sufficient to make the exchange of some advantage to her.

(2) Let the figures now be shifted in such manner as to conform to terms of trade under which 10 wheat exchange for more than 13 of cloth. For simplicity, we keep the United States figures as they were before, as regards money wages and the domestic supply prices of goods, confining the readjustments to the other countries.

Wages have now fallen in England from $1.40 to $1.10, and in Germany from $1.21 to $1.00. The supply prices of English and German goods have fallen correspondingly. Such is the nature of the results to be expected if a change in demand sets in which causes the barter terms of trade to be more favorable to the United States—if more wheat were demanded by England and Germany under the price conditions of Case 1 than was equal in money value to the cloth demanded under those conditions by the United States. Wages and the supply prices of goods are lower in England and Germany than they were before. The domestic supply price of wheat is now the same ($1.00) in Germany as it is in the United States, and wheat would not move between the two. But the price of wheat is lower than its domestic supply price in England ($1.10) and England would import wheat from both Germany and the United States. The domestic supply price of cloth, on the other hand, is lower in England ($0.73) than it is in either Germany ($0.77) or the United States ($1.00), and cloth would move from [106] England to both. The price of wheat thruout the trading area would be $1.00, the price of cloth $0.73. The terms of trade in physical units would be the ratio of those figures, that is 10 of wheat for 13.7 of cloth—more than 13 of cloth for 10 of wheat. And Germany would no longer be an exporter of cloth to the United States, but an exporter of wheat to England.⁰¹⁹

(3) Readjust finally in such way that the money wages and the supply prices correspond to barter terms of 10 wheat for 13 cloth. We still keep wages and prices in the United States at the original figures, confining the shifts to the other countries.

Under these conditions England and the United States exchange wheat and cloth; since the supply price of wheat in the United States ($1.00) is lower than that of wheat in England ($1.15), while the supply price of cloth in England ($0.77) is lower than that of cloth in the United States ($1.00). English cloth will be sold in the United States and England at $0.77, and American wheat will be sold in both at $1.00. The barter terms of trade will be 10 wheat for 13 of cloth (the ratio in physical units of the price relations $0.77 and $1.00).

Germany, however, can find no advantage from participation in trade on these terms. If indeed Germany and the United States alone were confronted with each other, trade would arise between them. Tho wheat is at the same price in both ($1.00), cloth is at $0.77 in Germany and at $1.00 in the United States; cloth would [107] move from Germany to the United States, specie at first move from the United States to Germany; in the end Germany would exchange cloth for American wheat. But English cloth already sells in the United States for $0.77, and Germany's supply price is that same figure—$0.77. Germany can gain nothing by the export of cloth to the United States, and the United States can gain nothing by the export of wheat to Germany. The barter terms of trade between England- and the United States are 10 wheat for 13 cloth. This is precisely the domestic term of trade within Germany between wheat and cloth; domestic supply prices in Germany are adjusted to this situation. England and the United States find it advantageous to trade on these terms; and so long as they do so on these precise terms, Germany has nothing to do with either of them.⁰²⁰

[108] So far those transactions only have been considered which arise out of sales and purchases of merchandise. Imports and exports of goods have been treated as if they constituted the sole operations in international trade and as if they alone gave occasion for international payments and the transfer of money. As is familiar enough, there are other operations of large consequence. Payments arising from international indebtedness—the making of loans and the payment of interest on loans—are perhaps the most important among them; most important because, for several generations at least, they have played a considerable part in the trade of many countries and over long stretches of time. Other payments also, for expenses of tourists, for charitable or family aid, indemnities payable after defeat in a war, have been important ; and tho less constantly in evidence than the items arising from indebtedness, they have at times risen to a commanding position. Charges for freight and passengers carried in the vessels of another country, and banking and insurance charges, are also substantial in amount.

These various transactions have come to be of increasing importance since the early part of the 19th century. It is true that all of them taken together have never been as large as the transactions on merchandise account. In no country and at no time—so far as I know—have they been equal (measured in terms of the sums of money involved) to the sales of goods between countries. But they have become large and have tended to constitute a growing proportion of the total. We may proceed to consider the principles applicable to them and the modifications of our main [109] conclusions which they suggest. The reader will bear in mind that it is the principles alone which are here to be considered. The actual operations will be described and discussed at some length in the second part of this book, and will give occasion for considerable qualifications, perhaps modifications, of the principles.

"Invisible" is the adjective commonly used to describe these items. They are invisible simply in the sense that they are not recorded as publicly as the imports and exports of merchandise, and on the whole are not so accurately known. The term is convenient when one wishes to speak of the whole series of items and to compare them with the goods transactions.

Begin with the simplest case of all: a remittance that has to be made from one country to another, with no quid pro quo obtained or to be obtained from the country receiving the payment. Such, for example, would be the remittance of income to absentee landlords ; more strikingly, a war indemnity payment, or a mere tribute. When there is, immediately or ultimately, directly or indirectly, a return of some sort by the receiving country—as with freight charges, tourist expenses, loans—the situation is different in some essential particulars. Eliminate this perhaps complicating element by taking a case in which there is no quid pro quo of any kind: something in the nature of a tribute.

Suppose further that a stated payment is to be made regularly year after year. Sporadic payments are commonly effected, under the modern organization of money and credit, by methods which disturb the ordinary course of trade to a surprisingly small extent; and very heavy payments of this kind are often settled with great smoothness. Steadily continuing payments, however, even tho moderate in amount, are not wound up without affecting the main current of international trade—the movement of goods from country to country.

Recall the figures already considered. Suppose:

[110] The United States has a comparative advantage in producing wheat, and will import linen from Germany even tho her labor is more effective in producing linen than is German labor. The barter terms of trade will be 10 of American wheat for anything between 10 and 15 of German linen. The nearer it is to 15 linen—the more linen Germany gives for 10 of wheat—the more the United States will gain; the nearer it is to 10 linen—the less linen Germany gives for 10 of wheat—the more Germany will gain.

Carry the case out in terms of prices and money wages. I select, as the starting point in the present set of illustrative figures, a situation in which the barter terms of trade are 12-1/2 of linen for 10 of wheat—that in which the gain from the trade is equally divided between the two countries.

The money cost of wheat is less in the United States than in Germany, and wheat moves from the United States. The money cost of linen is less in Germany, and linen moves thence to the United States.

Suppose, lastly, that at the prices stated ($0 85 for wheat and $0 68 for linen), the quantities of the two commodities that move are:

10,000,000 wheat exported from the U S at $0.85 = $8,500,000 12,500,000 linen exported from Germany at $0.68 = $8,500,000

The money amounts balance. Foreign exchange is at par; no specie flows. 10 millions of wheat exchange for 12-1/2 millions of linen; the barter terms of trade thus are 10 wheat for 12-1/2 linen.

Suppose now that the United States has to remit to Germany a million dollars annually—a tribute, or the like. The total money sum payable by people in the United States to those in Germany is now nine and a half millions—eight and a half millions for the [111] linen bought, and the tribute of one million. Foreign exchange is no longer at par; the American exports of wheat no longer yield bills on Germany in amounts sufficient to supply the needs of those who have to remit to Germany. Exchange on Germany rises to a premium in the United States; specie flows to Germany. Prices and money wages fall in the United States, rise in Germany. These changes will go on until a stage of equilibrium is reached, which may be exemplified as follows:

Wages have fallen in the United States from $1.70 to $1 60; they have risen in Germany from $1.02 to $1.05. The money cost of wheat has fallen in the United States from $0 85 to $0.80 and is now considerably lower than the German money cost of wheat; the money cost of linen in Germany has risen to $0 76f and is now not much lower than the American money cost of linen.

At these prices suppose the movement of goods to be

The exports of wheat from the United States exceed in money value the exports of linen from Germany. The difference is a million dollars, precisely the sum which has to be remitted to Germany; that is, it suffices to yield the volume of bills in Germany which are wanted by those persons (private individuals or public officials) having the remittance in charge. The demand for bills is just met by the supply; foreign exchange is at par; equilibrium has been reached.

The equilibrium, it is to be noticed, is one in the "balance of payments," not one in the "balance of trade." The payments to be made by the United States to Germany are completely met. But the balance of trade—the balance of merchandise operations—is "favorable" to the United States; her exports of goods exceed [112] her imports in money value. The balance of trade is "unfavorable" to Germany; her imports of goods exceed in money value her exports. These expressions "favorable" and "unfavorable" balance, with their implication that a country secures a gain in the one case and suffers a loss in the other, are so commonly used in the ordinary talk about international trade that it is difficult to keep away from them entirely. They rest on the obvious fact that if there be no other than merchandise transactions, an excess of exports over imports will cause a flow of specie into a country; and they rest further on the persistent mercantilist notion that there is something advantageous or "favorable" to a country in a relation of exports to imports which, if it stood by itself, would cause specie to flow in. The qualifying if has become of more and more importance in modern times, and consequently the mercantilist terminology, misleading in any case, has lost its significance even as a description of the forces on which depends the movement of specie. Non-merchandise transactions have become so large, and affect so steadily the trade of each and every country, that the relations of imports and exports in themselves give a very uncertain clue to that which in reality determines the specie flow. It is the balance of international payments which determines this flow. In our supposed case the balance of payments is precisely settled; the balance of trade, even tho it be called "favorable" to the United States and "unfavorable" to Germany, leads to no movement of specie either way.

In the important sense, the situation has become less favorable to the United States; and this in two ways. Not only do the people of the United States part with a considerable volume of tangible goods (wheat) in order to make the required payment to Germany, but in order to carry out the transaction and at the same time pay for the linen which they continue to buy, they ha\e to barter their wheat for linen on less advantageous terms.

Consider the figures Before the tribute became payable, the United States sent 10 millions of wheat to Germany, and got in exchange 12-1/2 millions of linen; for each ten of wheat 12-1/2 of linen were got. When the new equilibrium is attained and the annual [113] payment of the tribute is effected, the United States sends10-1/4 millions of wheat and gets but 9.4 millions of linen. She sends more wheat and gets less linen; she exchanges 10-1/4 of wheat for 9.4 of linen, i.e. 10 of wheat for 9.2 of linen. The barter terms of trade are much less favorable to the United States, much more favorable to Germany.

These figures, however, call for further consideration. The wheat sent from the United States is to be regarded as making two payments: one to meet the obligatory remittance, the other for the German linen. The two may be separated in this fashion:

The wheat that serves to pay for the linen amounts to 9,000,000 bushels. It is this quantity—less than the total sent—which can be said with accuracy to be exchanged for the 9,400,000 linen. The barter terms of trade, so considered, are 9 of wheat for 9.4 of linen, i.e. 10 Wheat for 10.4 linen. This is not so unfavorable to the United States as the relation just mentioned—10 for 9 2. But it remains much less favorable than the ratio of 10 to 12-1/2 which prevailed at the outset. To repeat, the people of the United States suffer loss in two ways. They send wheat to pay the tribute; and, in order to get the linen they want, they must give more wheat for each unit of linen which they continue to buy.

There are thus two ways of looking at the barter terms of trade. One may be indicated by the phrase "gross barter terms of trade"; the other by "net barter terms of trade." The first regards the whole volume of goods, both imports and exports. The second regards those goods only which pay for goods; it demarcates any movement of goods which serves for other payments. (I neglect services, for reasons presently to be explained.)

The gross barter terms in the present illustration are 10 wheat for 9.2 of linen; the net barter terms are 10 wheat for 10.4 linen. For some purposes the first is the important one, for other purposes the second. As regards the limiting figures—the range within which trade is possible—the net terms are alone important, [114] and indeed are alone to be considered. The net terms cannot be such that the United States gets for her 10 of wheat less than 10 of linen. But the gross terms may be disadvantageous to this drastic extent. The United States in the present case actually gets only 9.2 linen for every 10 of wheat she sends to Germany. True, she gets as much as 10.4 of linen for the money that is paid for every 10 of wheat she exports.⁰²¹ But she sends additional wheat to Germany, which serves for the tribute remittance and brings it about that on the total transactions she gets only the 9.2 linen for every 10 of wheat. And it is these total transactions which are really of significance for her welfare. Germany has no substantive concern in the manner in which the money account is drawn up and the balance of payments is reckoned, in analyzing how much of wheat is to be regarded as paying for linen and how much is to be set down for the other remittance. What actually happens is that the United States parts with 10-1/4 millions of wheat and receives no more than 9.4 millions of linen. Germany gets much wheat, gives little linen. And this situation persists indefinitely. Year after year—so long as the tribute continues and no other items enter—the United States sends to Germany a large slice of the product of her labor and receives a small slice of the product of German labor.

The degree to which the barter terms of trade, both net and gross, are altered to the disadvantage of the United States depends on the conditions of demand. The particular figures just chosen to illustrate the consequences of a tribute payment were such as would result from a play of demand unfavorable to the United States. They are conditions of inelastic demand in Germany for wheat and of elastic demand for linen in the United States; in more precise terms, conditions in which the elasticity of demand is less than unity in Germany and more than unity in the United States. Tho the price of wheat falls from $0.85 to $0.80, Germany buys but very little more wheat; and she spends on wheat a total sum less than she spent before—$8,200,000 now, $8,500,000 before. In [115] the United States, on the other hand, the amount of linen bought shrinks greatly in consequence of the rise in linen price (from $0.68 to $0.76-2/3); and the total amount which she spends on linen falls substantially, from $8,500,000 to $7,200,000. If the German conditions of demand were the opposite from these—elastic for wheat—there would be a mitigation of the American loss in the barter terms. The United States would still find that she exchanged wheat for linen on terms less favorable than before, but not so much less favorable as in these illustrative figures.

It need hardly be pointed out that in all such cases the figures of relative money wages are in the last analysis the results of the prices of the goods. They have been stated, for convenience of exposition, as if the wages determined the prices; the wages being the "supply prices." But it is the goods, of course, which first feel the impact of the play of international demand, and it is the prices of the goods which determine the money incomes. Wheat and linen rise or fall in price as changes take place in international payments; thence are derived the rates of wages; these wages then appear as the money costs, the supply prices, of the goods.

The reader who is not wearied by the details of such figures may follow them as they can be worked out for still one other sort of case, illustrative of a situation in which the play of demand is more favorable to the United States—that is, one in which the German demand for wheat is elastic (greater than unity) and the American demand for linen is also elastic.⁰²²

Reverting to the figures with which we started (p. 110), suppose once more an initial flow of specie, caused by a payment for tribute, and the consequent changes in prices and money incomes. Assume the following stage to have been reached:

[116] It will be observed that wages in the United States have fallen, but have fallen less than in the case just discussed: they have fallen from $1.70 (the figure at the original equilibrium) to $1.65, but not as low as $1.60. German wages on the other hand have risen, but not so much; they have risen from $1.02 to $1.05, not to $1.15. At the prices thus figured out, suppose:

At the new price of $0.82-1/2 for wheat (lower than $0.85) Germany takes a greater quantity—11 million bushels instead of 10 million; and her total payment to the United States for wheat rises from $8,500,000 to $9,050,000. At the new price of $0.70 for linen (higher than $0.68) the United States takes a less quantity of linen from Germany—11.5 million yards of linen instead of 12.5 million ; and her total payment to Germany falls from $8,500,000 to $8,050,000.⁰²⁴ The new gross barter terms of trade, i.e. the relation of all the United States wheat to the German linen, then become 11 wheat = 11-1/2 linen, or 10 wheat = 10.45 linen.

Of the total export of wheat from the United States, however, a part only serves to pay the tribute of $1,000,000; the rest pays for the German linen. The apportionment of the wheat exports for the two purposes is:

The net barter terms of trade are then:

9.8 wheat = 11-1/2 linen, or 10 wheat = 11.8 linen

Under the previous supposition, the net terms had been:

10 wheat = 10.4 linen

The United States now gets terms more favorable than before—11.8 of linen instead of 10.4 linen.

[117] Obviously, however, the United States gets terms which still remain less favorable than they were at the outset. We started, it will be remembered, with barter terms of trade such as to divide equally between the two countries the possible gain from the trade—10 wheat for 12-1/2 linen. The new terms just worked out, tho more favorable to the United States, are still not so favorable as those from which we began: the United States still gains less than at the outset.

The terms will be again shifted, and in the same direction, if we suppose the German demand for wheat to be still more elastic, and the American demand for linen still more elastic. All the figures will then be correspondingly modified—higher price of wheat, lower price of linen, higher money rates of wages in the United States, lower money rates of wages in Germany. But the rates of wages in the United States will always be lower than they were before the tribute payment set in, the German rates always higher. The barter terms of trade, again, might be but little less advantageous to the United States than before; they might be almost as much as 12-1/2 linen got for 10 wheat; but they would never be quite so much. The barter terms might be 12.2 or 12.4 linen for 10 of wheat, but so long as the annual remittance of $1,000,000 was necessary and no other new factor intervened, could never be 12.5. The United States always would have not only to pay the tribute, but would have to exchange its exports for its imports on less favorable terms.

"Less favorable terms." A distinction is to be drawn with regard to the significance of this designation according as it is applied to the sort of situation here analyzed, or to that arising from a mere change in demand. The latter case, that of a change in demand, was considered in a previous chapter.⁰²⁵ It was there pointed out that a change in demand is a voluntary act, or rather change of attitude, on the part of one or both of the exchangers. When the demand schedule, for example, shifts in such manner that at the same price more of a commodity is bought than before—if the demand curve moves to the right—the change means that people [118] are ready to pay a higher price for a given quantity, simply because they get greater satisfaction from that same quantity than they got before. True, they give more of their income for each unit of the commodity than they gave before, and in that sense may be said to buy it on less favorable terms. But there is no hedonistic loss; merely the registration of a different state of mind. And similarly when the people of one country choose to give more of their own goods in exchange for those of another country, the barter terms, stated as physical equivalents, become less favorable; but they do so merely because wants have changed and are now satisfied in a different way.

The case of a tribute is different. Tho the demand schedule in the tribute-paying country remains unaltered (as was assumed in the first part of this chapter) the barter terms become less favorable. Whether we look at the gross or the net terms, the tribute causes it to give more of its own goods, unit for unit, in the exchange for the goods of the other country. It can not console itself, as in the other case, by the reflection that after all this is precisely what its own changed state of mind has brought about. The only consolation is that it still gains from the trade. The goods which it sends out in payment for its imports still cost it less labor than would be needed for producing the imported goods at home. True, less of the imports are got in this exchange than would have been got if there were not the extra payment. But to continue the trade is the best way out of a bad business.

The case of a tribute is extreme; and for that very reason it has here been examined. It stands for a pure payment without quid pro quo. At the opposite extreme are payments, also standing for "invisible" items, where the country to which the payments are made and to which the additional goods flow does give in exchange something, even tho not visible goods; where there is clearly a quid pro quo. Commonly enough, in the discussion of this aspect of international trade, all the invisible items are lumped together, as if all had the same meaning and the same effects. Not so; they differ. To take a case analogous to a tribute, and nowadays familiar, reparation or indemnity payments stand for one sort of effect; whereas payments [119] which arise from the expenditure of tourists in foreign countries stand for quite another. Payments connected with foreign loans stand midway; and this whether we consider the initial lending of the principal amount by the creditor country or the subsequent payment of interest by the debtor country. I shall say something of loan and interest payments in another connection.⁰²⁶ For the present, by way of elucidating the essential differences between the several sorts of cases, we may consider a case which stands at the opposite extreme from tributes or indemnities—that of tourist expenditures.

The expenses of Americans who travel abroad form a large item in the balance of payments of the United States. They give occasion to remittances to foreign countries, and, thru the process just explained, tend to cause merchandise exports to exceed imports.⁰²⁷ This item, if it were the only transaction other than sales of goods—the only invisible item—would bring about a balance of trade "favorable" to the United States. As regards the physical goods exported and imported, the situation of course would not be favorable to the United States; the relation of American incomes to foreign incomes, and the barter terms of trade would become less advantageous to the United States than before. So far the case is the same as with a tribute.

Obviously, however, there are differences. In return for the additional commodities exported—the excess of exports—the Americans get not indeed imported goods, but the pleasures of travel. Taken as a body, they prefer these pleasures to the enjoyments which would have been yielded by the exported goods, or by their equivalents, if consumed at home. To state the same thing in another way, the American tourists, by spending abroad, cause American labor to be turned to making exported commodities, rather than to making such commodities as the travellers would have purchased if they had remained at home. There can be here no question of a loss, such as a tribute would entail; it is merely a [120] matter whether expenditure in one direction is preferred to expenditure in another. The American demand schedule has shifted. Regarding the tourists as one among the various groups of Americans who find foreign products to their liking, the American people as a whole now want more of foreign things than before. A readjustment of the barter terms of trade is necessarily involved; but it no more involves a real loss, in the hedonistic calculus, than any case of change in demand. It is a matter of what people prefer.

Another point may be raised: the relation of such remittances to the distribution of wealth within a country. In the preceding paragraphs it has been tacitly assumed, for simplification of the problem, that the Americans, travellers and stay-at-homes, are a homogeneous set of persons. We have neglected what is suggested by the familiar conditions of travel—that, so far from there being homogeneity, the travellers are the rich, while the bulk of the Americans and the main consumers of imports are those of slender means. We have supposed, then, that the American travellers are a sample of the Americans as a whole; or, what amounts to the same thing for the purpose in hand, that the set that travels is the identical set that is buying the imports. Applying this supposition to our illustrative case, the American travellers and the American purchasers of German linens may be regarded as the same group. Then these travellers not only have to meet their expenditures abroad, but have also to face the fact that their German linen bought in the United States is more expensive than before, while their money incomes (derived from domestic sources) are smaller than before. If they nevertheless continue to travel, it must be because the attractions are so great as to outweigh all the drawbacks, increased expense of linen included. The net gain in satisfactions or gratifications remains; otherwise this particular choice would not be made.

It is to be granted, of course, that the supposed homogeneity of purchasers is not necessarily, perhaps not generally, in accord with fact. The American tourists may be quite a different set of persons from those who buy the goods imported. But this consideration introduces an extraneous set of factors—the distribution [121] of income in the United States, its inequality between classes or geographical sections, its possible disturbance and readjustment under new conditions. Any changes in the direction of consumption may have effects not only on the direction of production but on the distribution of income also. This problem involves reasoning which is in part similar to that on international trade, but leads to nothing inconsistent with its conclusions or serving to modify the essentials of the conclusions.

A somewhat special case is that of gifts or charitable contributions. It is like that of a tribute, and also unlike. As with a tribute, nothing in the way of commodities or services is received by the country which makes the payments: there is no quid pro quo. But they are made voluntarily, not under compulsion. Very heavy remittances of this sort were made from the United States to foreign countries during the last generation, say from 1895 to the present time (1925); some details will be considered in a later chapter.⁰²⁸ Such operations affect the course of international trade in commodities—the exports and imports of goods—in the same way as tributes or travellers' expenditures. Their tendency is to bring about an excess of merchandise exports, a "favorable" balance of trade, lowered money incomes and domestic prices in the remitting country, higher incomes and prices in the receiving country. They lead also to barter terms of trade less favorable to the remitting country. The people of that country not only export gratis the tangible goods which serve to meet the charitable remittances; they lose also thru the circumstance that they get less imported goods in exchange for the exports which are the commercial items in the account. The remittances thus may be said to cost, the donors more than they reckoned on, more than they are aware of. But the contributions continue to flow, even tho the people who make them find that their money incomes tend to fall, while imported goods tend to rise in price. They have the satisfaction, approved by the moralist, of doing a merciful deed; and that satisfaction is not dimmed because the doing entails more of material curtailment than was resolved on at the start. On the [122] principles of a higher or sublimated utilitarianism, they suffer no loss, nay, reap the highest gains, thru the whole gamut of the performance.⁰²⁹

[123] LOANS made by the people of one country to those of another, and interest payment on such loans, have already been mentioned as among the most important of the non-merchandise items in international trade. They also, while having effects similar in the main to those of other invisible items, present some problems of their own.

For illustration we may suppose that loans are made by British to Americans. For brevity, we commonly speak of such loans as made by Great Britain to the United States, as if one government made them to the other, or the British as one body or entity made them to the Americans as another. In fact, the transactions are commonly between individuals, or (what comes to the same thing for bur purpose) between individuals on one side and political bodies on the other. Loans are indeed sometimes deliberately made by one state to another; such operations played a large part in the Great War of 1914-18, and were not unknown in earlier periods They are the results of political or military exigencies, and while involving no principles different from those applicable to the transactions between individuals, are yet likely to have a range and scope quite beyond those of ordinary commerce. For this reason they will be considered separately in later chapters. Here we confine attention to loans by individuals, not of an emergency or catastrophic sort, made for profit, exercising their effects gradually and as a rule quietly on the every-day phenomena of international trade.

Such loans by the one party, borrowings by the other, must result in a flow of specie from Great Britain to the United States. "Must result"—this puts the case too strongly. The flow will [124] not necessarily take place; possibly there will be none at all. And such flow as does take place is not likely to be equal in volume, either in the very first stage or later, to the amount of the loan. And yet it can be said almost with certainty that some specie movement there will be.

The possible but improbable case where there will be no movement of specie at all is when the borrowers use the entire amount of the funds put at their disposal by the lenders, in buying commodities in the lenders' country. And further: the borrowers not only make purchases in the lenders' country, but these purchases are additional to what would have been made in any event. Suppose the lenders, for example, to be British, the borrowers Americans—the sort of relation which existed between these two peoples thru the 19th century. Suppose the Americans are railway promoters who use the entire proceeds of the loan in Great Britain for buying rails, locomotives, bridge material, and the like. Other American purchases go on as before, and other goods continue to move from Great Britain to the United States as before. The new purchases and new exports exactly absorb the funds which British lenders have put at the disposal of American investors. No remittances at all will be made from Great Britain to the United States. English commodities will go to the United States as the direct result of the loan.

This sort of consequence—an immediate export of goods from the lending country, and for the time being no further change—may ensue as the result either of the ordinary economic forces, or of a set policy in which there is deliberate or conscious diversion of international trade. In the period since 1890 there has been much endeavor of the second kind. This was often the case in France and Germany during the generation preceding the Great War of 1914-18. It was the undisguised policy of the governments in both countries, and of the financial promoters and institutions which were in close touch with the governments, to arrange the terms of foreign loans in such way that the borrowers should spend the entire proceeds in France or Germany. Virtually the same sort of thing appeared in the huge loans which were made from the United States [125] to the Allies during the Great War itself; tho here, as will be shown in a later chapter, the conditions were quite exceptional and the consequences unusual. It appeared again, and under conditions not so exceptional, in the American loans of the post-war years. It may continue to play a considerable part in the future. The bankers who float loans are often representatives of manufacturing enterprises for whose output they wish to secure a market. Governments and the business public are fairly obsessed with a determination to promote exports in every possible way—the ineradicable spirit of mercantilism. And where the loans are made not merely for industrial purposes, but for military or naval equipment, the combination of political and economic motives acts even more strongly to link foreign loans directly with commodity exports.

It is the other sort of interlinking, however, that not deliberately designed, which has played the larger part in the past and may be expected on the whole to do so in the future. During the greater part of the 19th century loans were made without express stipulation of the kind just described. Great Britain was then the main lending country. Great Britain was also the cheapest place in which to buy industrial equipment. Borrowers laid out a portion of the borrowed funds, tho not often the whole, in buying British goods; they did so merely because they found it to their own advantage to do so. The same has been the situation with most of the loans made by the United States to foreign countries in the post-war period, or at least after 1920 The borrowers are free to do as they please with the proceeds of the loans, and it is not to be foreseen whether they will use them in any part for purchases in the United States.

In all these cases, whether there be express stipulation concerning the purposes to which the loans shall be devoted, or a purely commercial use of the funds in the lending country, the effect of the borrowing on the substantive course of international trade becomes direct. The merchandise movements and the merchandise balance of trade are affected at once. Merchandise exports from the lending country exceed merchandise imports, without any intermediate stage of disturbance of the foreign exchanges, [126] flow of specie, and so on. The balance of trade becomes at once "favorable" to the lending country, and "unfavorable" for the borrowing country. There is no disturbance of foreign exchange, no flow of specie, nothing to modify the level of prices or wages either in the lending or in the borrowing country.

It is extremely rare, however, that the purchases of goods in the lending countries by the selfsame foreigners who contracted the loans take place to such an extent as to obviate the flow of specie completely. Not the entire proceeds of loans are likely to be spent in this way, only some fraction. Even if railway promoters from the United States or Canada or Argentina, who borrow in England, also buy railway material in England, they are likely to use in this way only a part of the funds. Some part they will spend at home, for labor, for miscellaneous supplies, divers expenses. It is conceivable, nay probable, that they will raise some portion of their capital at home, and only the residue abroad. And it is then conceivable that they will use for domestic expenditures the funds raised at home, and will use the proceeds of foreign loans entirely for purchase abroad. But it is most improbable, even when there is a division between foreign and domestic financing, that an exact balance of this sort will be struck. In the majority of cases a part of the foreign funds, and usually a considerable part, will be wanted for expenditure in the borrowing country itself. Then, to repeat, the outcome must be a flow of specie from the lending to the borrowing country. Remittances will have to be made, in our illustrative situation, from London to New York. Specie will flow; the consequences become the same as those which ensue when remittances have to be made for any other invisible item.

These consequences will of course ensue quite without modification if there be no immediate purchases of goods at all in the lending country. Such was doubtless the case with a large proportion of the British loans both of earlier and later date. It was so with the continuous stream of loans by the French in those earlier loans of the second and third quarters of the 19th century, made when neo-mercantilism was not yet rampant. The transactions were [127] such as to involve at the outset nothing more than the obligation to put funds at the disposal of the borrowers; while the borrowers themselves transferred these funds, except for the possible use of some fraction forthwith in the lending country, to their own country. All in all, we are justified in treating this as the normal and ordinary course of events. International loans disturb the existing balance of payments; remittances are made to the borrowing country; specie flows thence from the lending country.

The further course which events may then be expected to take is sufficiently familiar, and need not again be analyzed in detail. The loan being made (in our assumed case) by British to Americans, prices and incomes fall in Great Britain, rise in the United States. An excess of exports develops in Great Britain; not immediately, but by a gradual process. She comes to have a "favorable" balance of trade. In the United States an excess of imports gradually appears—an "unfavorable" balance of trade. The people of Great Britain send merchandise to the United States, and add to the tangible equipment of the Americans, or to their consumable goods, giving up for the time being some of their own possessions and adding to those of the Americans. But not only do they give up something in this way—make a sacrifice, incur a loss, for the time being—but they incur a further loss in that the barter terms of trade become less advantageous to them. The imports which they continue to buy from the United States are got on less favorable terms than before. Conversely, the people of the United States have a double gain; not only do they get an extra supply of imported goods, but all the imports, the goods plainly and simply bartered as well as the extra goods that represent the loans, are got on better terms than before.

The ulterior consequences on the barter terms of trade, let it be repeated, will not appear so far as the borrowers make direct purchases of goods in the lenders' country. And if the borrowed funds are used in toto for such purchases, the ulterior effect will not ensue at all. If part is so used, the effects will be mitigated. The actuating machinery for these effects is the flow of specie, which is eliminated so far as there are the direct purchases.

[128] Proceed now a step further. Assume again a simple case, for elucidation of the principles. Suppose that loans go on year after year, the same amount annually. Each year Englishmen lend to Americans a given sum, say 10 millions. Assume also, for simplicity, that there are no direct purchases by the borrowers, but always—in the first instance, that is—a flow of specie into the borrowing country. In due time, the length of the interval depending on the sensitiveness of prices to the increase or decrease of specie, a continuing favorable balance of trade appears in Great Britain and the reverse appears in the United States. The lending country has a steady excess of exports, the borrowing country a steady excess of imports. Specie no longer flows; the continuing loans are made thru the mechanism of merchandise movements.

At an early stage in the operations, however, another factor begins to enter. Each year the borrowing country has to pay interest on the loans contracted so far; and to that extent the amount which the lending country has to remit on capital account is reduced, as regards the net balance of the international account. The interest charge to be paid by the borrowing country grows with every year. The capital sum from the lending country remains (under our supposition) the same from year to year. The accumulating interest charge will grow, and in time will be equal to the constant capital sum. Eventually, it will be greater. At the outset the transactions lead the lending country to make remittances to the borrowing; in the end it is the borrowing country which has to remit.

These shifts in the relations between creditor and debtor country will manifest themselves in the flow of specie between them and in their merchandise transactions. The initial flow from the lending country is destined to be checked in any case by the changes in prices. But it will be checked the more quickly by the accruing interest charge. The accommodation of the merchandise balance (the balance of trade) to the balance of payments will therefore take place more promptly than in the case of other invisible items, such as tourist expenditures. And eventually there will be a reversal of the initial features. Specie will flow back to the [129] lending country; its prices and money incomes will rise; the borrowing country will have falling prices and incomes; the lending country will come to have an excess of merchandise imports and the borrowing country an excess of merchandise exports. A cycle of operations is set in motion by a steady succession of international loans. Their effects on international trade and international payments are different according as the transactions are in the initial stage, the midway stage, the final stage.

In popular talk on these matters it is commonly assumed that a creditor country ipso facto has an excess of merchandise imports, and a debtor country an excess of exports. The creditor country—so people imply in their everyday talk—has payments to receive, the debtor country has payments to make; the former is expected to show a net credit in its accounts, the latter a net charge or net outgo. Not at all. The state of the merchandise account, the balance of the money values of imports and exports, may run either way, for either debtor country or creditor country. It depends on the stage which the credit operations have reached. And, similarly, there is a common erroneous notion that the flow of specie tends to be toward the creditor country; that the course of the foreign exchanges is naturally such as to cause specie to move toJt, or at least such as to bring some pressure that way. Again not at all. The movement of specie may tend to be in one direction or the other, according to the stage of the cycle.

It is to be remarked, however, that the transactions rarely show such regularity as the preceding analysis has implied. On the contrary, they usually take place with marked irregularities. And not only are they irregular; they are subject to abrupt stoppages. They frequently entail spasmodic changes in international payments and in the movement of goods.

These irregularities, of which abundant illustrations will be given in later chapters, deserve some further consideration even at this point. If loans on capital account were continued regularly at the same amount year after year, the accumulating interest payments would bring about, at the date when reversal of the relations began to set in, a slow and gradual readjustment, not a [130] sudden overturn. The borrowing country would almost imperceptibly accommodate itself to a new situation, in which specie would seep out, prices gradually fall, merchandise exports rise and imports fall. A "favorable" balance would become in time a settled feature of its international trade. In fact, however, the loans from the creditor country, so far from being made at the same rate year by year, begin with modest amounts, then increase, and proceed crescendo. They are likely to be made in exceptionally larger amounts toward the culminating stage of a period of activity and speculative upswing, and during that stage become larger from month to month so long as the upswing continues. With the advent of a crisis, they are at once cut down sharply, even cease entirely. The interest payments on the old loans thereupon are no longer offset by any new loans; they become instantly a net charge to be met by the borrowing country. A sudden reversal takes place in the debtor country's international balance sheet; it feels the consequences abruptly, in an immediate need of increased remittances to the creditor country, in a strain on its banks, high rates of discount, falling prices. And this train of events may ^ ensue not once only, but two or three times in succession. After the first crisis and the first overturn, the debtor country is likely to recover. Within a few years loans from the creditor country may be resumed, another period of activity and speculative investment set in, the old round repeated, until finally another crisis comes and another sudden overturn in the balance of international payments. The final outcome, when this long period of irregular movements has run its course, is that the debtor country has more to remit on interest account than to receive on principal account, and that the remittance is effected by an excess of merchandise exports over imports. The history of the United States and of Argentina, both of which were typical borrowing countries at similar stages in their economic development, shows these successive waves of international borrowings, repeated crises, deviations from the simplified process set forth in the preceding pages. On the whole and in the long run, the actual course of events conforms to the theoretic analysis. The consequences indicated [131] by that analysis, so far from being completely obliterated by the irregularities, rather become accentuated and more conspicuous. The unmistakable fact of experience is that a country which is in the early stages of lending to others has an excess of merchandise exports; it has a "favorable" balance of trade. On the other hand, a country in the early stages of borrowing has an excess of imports—an "unfavorable" balance. At the further end of the international credit cycle, a country which for decades and generations has been making foreign investments, and to which, therefore, interest payments have been steadily accumulating, has an excess of merchandise imports, an "unfavorable" balance. Conversely, a country which is in the early stages of borrowing does in fact have an excess of imports; but when it has been a borrower over a long period, it has an excess of exports. If the borrowing process has ceased, or has greatly declined, this stage is the more pronounced. It is reached, if its borrowing operations have already gone on for decades and generations, even in the face of continuing large loans. With all the irregularities in the steps by which the successive stages are traversed, the stages themselves are in almost every case to be discerned; and they follow one upon another in the order which general reasoning leads us to expect.

Reverting now to that part of the theoretical analysis which relates to the barter terms of trade, the reader will observe that while these terms tend to be made more favorable to the borrowing country during the earlier phase of the cycle, they become in the later phase less favorable to the borrowing country and more favorable to the lending. What the borrowers as a people lose at the start, they are likely to regain at the end. Obviously, the balancing of loss and gain is not likely to be precise; least of all is any such offsetting to be clearly discerned or measurable. Whether there proves to be in the end a final surplus of gain one way or the other will depend on the demand for the commodities of each country by the people of the other. Both as regards the commodities exchanged and their demand schedules, there may easily be changes during the long period—a generation, a half-century—[132] over which the whole series of operations extends. All that can be said is that there is a general off-setting tendency between the earlier and the later stages, and a presumption that in the entire balance of this account neither country is likely to have any considerable net gain.⁰³⁰

Freight charges and passenger fares constitute another important item. Of the two, freight charges are the larger, and while in the main they raise no new questions, some aspects deserve separate attention.

The item of freight charges appears more particularly in the trade between countries separated by large stretches of ocean. Here there is a considerable gap between the place where the goods leave one country and that where they enter another; and a charge arises which does not necessarily form a part of the expenses of production within either. It is this circumstance—that the expenses may be incurred by residents of either country and that payments therefore may become due either way—which is peculiar to the item of freight charges. In other respects they present no peculiar features. They constitute a payment for service rendered ; and so far as the service is rendered to persons in one country by persons living in another, payments must be made to the foreigner. The item figures in the international balance sheet like any other, with the same effects as a payment for goods. It is invisible, too, in the same sense as tourist expenses are; that is, no official record is made and the amounts that must be remitted are often not easily ascertained.

There is a distinction, however, between the sum which the individual purchaser of foreign goods must pay for them, and the amounts which must be remitted to the foreign countries in payment for the goods. The individual purchasers must pay the foreign [133] price (i.e., the price at the place of export) plus freight charges. But this is not necessarily the amount which the people of the importing country have to remit to the exporting country, say Great Britain. If the goods are carried from Great Britain in vessels of the United States, the freight charges are paid by one set of Americans to another set of Americans. The freight item then is purely domestic; no remittance to Britain must be made. The British goods alone need to be paid for. But if the goods are carried in British vessels, some British persons must be paid for the further service of bringing them over. It is immaterial to the individual Americans who happen to buy the goods whether _sthis additional payment goes to their own countrymen or to the British. But it is material for the balance of international payments; an item arises in the latter case which must take its place in the adjustment of that balance.

The same distinction of course must be made at the other end—that is, as regards a country's exports. American sellers of goods, when they export them to Great Britain, get only the price of goods at the place of export. The British purchasers pay as individuals that price plus cost of ocean transportation. If the goods are carried in American vessels, the freight charges become an additional item, also payable to Americans, even tho (in modern times) presumably a different set of individuals from those that have sold the goods. Should the goods be carried in British vessels, the freight charges become a payment made by one set of the British to another set, not made by the British to the Americans; and then it does not figure in the international balance sheet.

If, now, the business is halved—if half of the carriage is done by British vessels, half by American—the items offset each other in the international account. As much is due one way as the other. The total volume of international payments is greater than it would be if the countries were contiguous, but the balance of payments is not affected. If, however, all the carriage takes place in the vessels of one of the countries, say Great Britain, a balance becomes due to that country. Supposing the other transactions between the countries to balance—imports and exports, and [134] whatever further items there may be—the additional sum due to Great Britain will be provided in the same way as in other cases of unsettled balance of payments. The theoretical solution is familiar. If payment had balanced before this item was present—if we suppose this to appear as a new item—specie flows to Great Britain; a double set of price changes sets in, upward in Great Britain, downward in the United States; imports and exports are modified; finally Great Britain has an excess of merchandise imports, the United States an excess of merchandise exports. And this series of changes brings the familiar consequences for the barter terms of trade; they become less favorable to the United States, more so to Great Britain. The people of the United States get their imported goods on less favorable terms than before; those of Great Britain get theirs on more favorable terms.

Shipping charges and shipping earnings thus have a place in international trade precisely like that of tourist expenditures. They take their place in the balance of payments, and they affect the net barter terms of trade—these only, not the gross terms. The Americans (say) pay the freight charges to the British, and get the freight service; they get their quid pro quo at once. They pay others for doing the work of carriage, rather than do it themselves; and the reason why they make the payment, in the last analysis, is that the others can do the work of carriage cheaply, while they themselves can apply their labor more effectively in other ways. It is quite superfluous to explain, to those who follow the general reasoning of the theory of international trade, that there is no net loss to the Americans from their payment of shipping charges. The case is similar to that of tourist expenditures and dissimilar to some of the other transactions considered in the preceding pages, in that there is an immediate service, for which payment is made at once. At once, that is, in the sense in which it can be said that the entire balance of international payments is settled at once; it is a balance settled very promptly, within a few months or a year. Of the temporary extensions and adjustments of "unfunded"balances more will be said elsewhere; they cause no modification of the general principles here under consideration. [135] Freight charges, to repeat, constitute items in the international account, essentially like the purchases and sales of merchandise, and are settled as promptly as these. The mere payment of them no more constitutes a source of loss to the paying country than does its payment for imported goods.

The possible effects of transportation charges on the barter terms of trade was the occasion for discussions and distinctions which held a considerable place in the older literature of the subject and may be briefly mentioned. There is an obvious gap (as has been noted) between the sum which the exporter receives for his goods, and that which the importer pays for those same goods: the gap standing for the transportation or freight charge. The price which the importer pays, and which he then charges to the consumer, is higher than it would be if there were no freight charge at all. Consequently the amount which the consumer purchases will be different from what it would have been if the price had not been so raised. This reaction of price on quantity demanded takes place on both sides; in our supposed case, it takes place among the purchasers both in Great Britain and the United States. The British buy less than they would have bought if there were no expense of transport; the Americans likewise buy less. But the effect on demand will not necessarily or probably be the same on both sides. It is not likely that the elasticity of demand for imported goods is the same in the United States for British goods as it is in Great Britain for American goods. The barter terms of trade, then, under the interplay of mutual demands, will be different from what they would have been in the absence of transportation charges—different from what they would have been between quite contiguous countries. In this sense, and in this sense only, it can be said that freight charges do not necessarily constitute an unalloyed burden on the receiving (importing) country, but may be borne in part by the despatching (exporting) country; indeed, conceivably borne by this country in whole. The price of every imported article is higher to the purchaser in the importing country by the amount of the transportation charge; in this direct and obvious sense the charge is borne by the importing [136] country, not by the exporting. But in general price levels and income levels, and thereby in the barter terms of trade, there may ensue conditions different from what would have been in the absence of such a charge. In consequence the net gain from the exchange of commodities may be quite as great to one of the trading countries, or nearly as great, as if they had been contiguous; nay, it is conceivable, even greater than if they had been contiguous.

These, however, are recondite possibilities, quite beyond me ken of the individual buyers and sellers, and of interest only to the speculative economist. Even for him they constitute an intellectual plaything rather than a matter of substantive importance. They are no more than ramifications of the abstract theory, and belong among the phases of the theory which it is impossible to verify or illustrate from the actual course of events.⁰³¹

Of some substantive importance, on the other hand, is a complication which the item of freight charges entails in the interpretation of import and export statistics.⁰³²

The common practice in the compilation of official statistics on the movement of merchandise is not the same for exports as for imports. To put the difference in commercial terms, exports are usually figured f.o.b. (free on board), imports usually c.i.f. (cost plus insurance and freight). The exports, that is, are recorded in terms of the value of the goods as put on board at the place of departure; the imports are recorded in terms of the value of the goods as received at the place of arrival. When valuing exports, cost of transportation is ignored; when valuing imports, it is included. The consequence is that recorded imports tend to be over-stated to the extent of that item, while recorded exports itend to be understated.

The statistical puzzles to which the practice leads can be best illustrated by analyzing some possible cases. Simplifying the analysis, as has been done for other problems, we may take [137] the trade of two countries only. Suppose that between these two the shipping trade is equally divided, the goods being carried half in the vessels of one of them, half in the vessels of the other. The charges for shipping thus cancel each other in the balance of international payments, and in any calculation of what may be due from one country to the other they may be disregarded. Suppose now that the merchandise exports and imports also balance, and that there is thus a settled equilibrium in the total account. Nevertheless, under the usual practice, each country would show in its official statistics an excess of imports; each would apparently have to make a remittance to the other in order to balance the international account. The shipping charges are ignored in the statistics of both as regards carriage one way; but as they are equally divided between the countries, the omissions cancel. An international trade account which on its face seems to be doubly unstable, indeed incomprehensible, is in fact stable and simple.

Now suppose not that the shipping trade is equally divided between the two countries, but that all the carriage is done in vessels of one of them. Take for example Great Britain and Australia. Both keep their statistics in the usual way, but the carrying is all done in British vessels. The actual position of Australia in the international account will then be in accord with that shown by her trade statistics; but the actual position of Great Britain will not. Australia in fact has to pay Great Britain not only for the British goods imported, but for those goods as delivered in Australian ports, i.e. plus freight. She receives from Great Britain only the value of her own export f.o.b. (not including freight). If then her official statistics show an excess of imports (as in the case considered in the preceding paragraph), this indicates that in fact a balance must be paid to Great Britain, and specie must be sent in settlement. If, however, her exports as recorded, so far from being less than the imports, appear to equal them, there will be an established equilibrium. Only when the exports f.o.b. equal the imports c.i.f. will there be a settled balance of payments for a country which does no shipping of its own and keeps its official statistics on the usual plan.

[138] Great Britain, on the other hand, must show under these conditions (the carrying all done in British vessels) an accentuated excess of imports—a greater excess of imports than would appear on her records if the carrying trade were equally divided. She must have a real excess of imports of merchandise, not merely the nominal excess which the statistical practice of itself tends to show. Her recorded imports are doubly swelled; first by the practice of valuing them c.i.f., and second by the substantial fact that she has payments to receive for the carriage of her exports.

Not all countries, however, follow the practice of valuing exports one way, imports the other; and here the complication becomes different. The United States, for example, instead of valuing imports on the c.i.f. basis, values them as the exports are valued in most other countries; that is, on the f.o.b. basis. Her statistics give the values of imports at the time and place of exportation from the foreign country, not their values on arrival at the United States ports.⁰³³ Cost of carriage is ignored in the official records of imports as well as of exports.

Here again we may consider the same two representative cases: one in which the shipping trade is equally divided, the other in which it is all in the hands of one country. Take the United States and Great Britain, again, as Australia and Great Britain were taken before. If, first, the carrying trade is equally divided between the United States and Great Britain, and the charges on this account just balance each other, the official statistics will show, as regards exports and imports, an excess of imports for Great Britain but for the United States imports just equal to the exports—no difference either way. For the United States the official showing will be in accord with the facts—a settled equilibrium. But for Great Britain an excess of imports will be shown; and that excess, so far from indicating an unsettled balance, will be but nominal. Great Britain will not in fact have any payment to make to the United States; her imports will only seem to exceed her exports.

[139] If now the shipping is all done in British vessels (this of course was the sort of situation that prevailed as between these two countries for a generation preceding the war of 1914-18), the actual relatiqns become different. The official statistics will again reflect the change; but again with figures not indicative of the real situation. If the merchandise imports and exports of the United States, as recorded by the United States, are the same in money value—that is, if the imports f.o.b. just equal the exports f.o.b.—the United States has nothing left with which to pay the freight charges due to British vessel owners. In due time the relation of imports to exports will become such as to bring about payment for this extra item; the exports recorded f.o.b. must exceed the imports recorded f.o.b. by the amount of the freight charges. An equilibrium of international payments will be reached only when the United States statistics regularly show an excess of imports. And this will indicate the real situation: the United States will be paying for the shipping services by sending merchandise to Great Britain. On the other hand, Great Britain, whose records would show an excess of imports in any case, because of her statistical practice, will again show—as in the case of trade with Australia—a greater excess of imports than would be shown if the shipping trade were equally divided. Part of her import excess will be nominal, but part will be real.

In sum, the usual statistical practice—that of valuing imports c.i.f. and exports f.o.b.—makes the imports of most countries appear large in relation to their exports. If all countries kept their records in this way, all would tend to show an excess of imports. That is, to state it more carefully, if merchandise exports and imports were such as exactly to pay for each other—if this, the simplest situation in international trade, were established, and if shipping trade were equally divided, so that nothing from this factor intervened to disturb the simplicity of the situation—nevertheless the statistics would show for each and every country an excess of merchandise imports. And therefore if the shipping trade is not equally divided, allowance has to be made for this continuing deceptive circumstance; a discount, so to speak, has [140] to be made from the recorded imports. Where still other invisible items than freight enter in the international account, such as loans, interest, tourist expenses, and the like, this continuing deceptive circumstance must still be allowed for. The imports always appear too large, and some discount must be made. When, on the other hand, the usual practice is departed from, as in the United States, a qualification must be made as regards the allowance : the discount is to be applied to foreign nations as regards their recorded trade, but not to the United States as regards her recorded trade.

[141] IT would be possible, to carry much further the sort of analysis which has been undertaken in the preceding chapters. The fact that each country deals not with one other only, but with many, would lead to modifications or elaborations over and above those already considered. The competition of various countries as buyers and sellers obviously has the effect of limiting more narrowly the range within which the barter terms of trade may vary; the terms of trade become dependent not on the demand schedules of any pair or trio of countries, but on those of all the trading countries combined. So much goes without saying.

More intricate are the possibilities in another direction. The imposition of taxes on imports and exports may affect the barter terms of trade—almost surely will do so. A tax on imports is equivalent to a deliberate lessening by the taxing country of its demand for foreign products. A tax on exports is equivalent to calling on other countries to decide whether they will continue to lay out as much as before on the taxing country's products. These may be described as intentional deflections of the play of demand; and they may be analyzed as having different effects according to the way in which they are levied—whether as taxes in kind, or (of course the only way that signifies in practice) as taxes in money. The effects which taxes may have on the volume of international trade and on the barter terms of trade have been the occasion of some of the most ingenious and intricate theoretical reasoning, and some most remarkable manifestations of casuistic ability.

I shall not attempt to refine further in the theoretical analysis. In doing so, there is always danger of lapsing into intellectual gymnastics. The suppositions made are sometimes improbable [142] to the point of unreality (as, for example, that of taxes in kind). More commonly, the deduced conclusions, even if resting on probable assumptions, are such as cannot be specifically discerned or verified in the actual course of events. And, as I need hardly confess again, the mathematical processes by which alone some of these conclusions can be deduced are beyond my competence; I could make no pretense of contributing anything new either in substance or in the way of exposition.

Neither is it within the scope of the present volume to enter on the controversy regarding free trade and protection. This in its main outlines is simple; simple at all events as compared with the topics taken up in the preceding pages. In a later chapter I have summarized those results of my inquiries on the effects of tariff legislation which have some direct bearing on the principles with which the present volume deals.

There is, however, one possible effect of taxes, and one phase of the protective controversy, on which something may here be said. I direct attention to this particular point of theory because of its connection with certain concrete problems of verification or interpretation which arise in connection with the international trade of the United States. The point is not of an essentially new or intricate kind. It relates to the effects which taxes on imports, and especially taxes which are protective, may have on the barter terms of trade.

Suppose, first, that a country imposes duties on imports which are purely of a revenue character. The proximate effect is to raise within the country the price of the dutiable article or articles (hereafter we may speak for simplicity of but a single article). True, in the case of a commodity produced under monopoly conditions, and having an extraordinarily elastic demand schedule, the price might remain unaffected. But this is a negligible case; under almost every imaginable condition there will be some rise in price. Assuming then, as we may, that price rises, less of the commodity will be bought. Only if demand were absolutely inelastic would the quantity bought remain the same. Demand being always in some degree elastic, less will be bought, and imports will decline.

[143] Observe that while price rises to the consumer, the total sum which is paid to the foreign producer will not rise. The quantity bought from the foreigner—the number-of units of the commodity which are purchased from him—becomes less, while the price per unit paid to him will not rise. True, because of the duty the consumer will pay a higher price. But the addition to the price goes to the government; it has nothing to do with the price received by the foreign exporter. Imports will decline, not because a larger price has to be paid to the foreigner for the goods but because an addition to the foreigner's price must be paid in order to meet the tax imposed at home.

The extent to which imports then decline will depend on the degree in which the demand for the commodity is elastic—whether the elasticity of demand is less or greater than unity. If demand be inelastic, the consumer will continue to buy nearly the same quantity as before. Imports will decline but little. The total sum spent for the article by the consumers will be greater than before, and the government will secure a large revenue from the tax. If on the other hand demand be elastic, the consumer will be led to lessen his purchases of the article very considerably, and the total sum spent for it will shrink. Then imports decline heavily, and the revenue which the tax yields to the government becomes correspondingly less.

But, to repeat, in any event imports will decline somewhat. And that decline at once sets in motion a train of forces which diminish the volume of international trade and at the same time cause the barter terms of trade to be more favorable to the country imposing the duties. Suppose the tax-imposing country to be the United States. That country's imports for the time become less. Exports, however, remain as great as before. Specie flows in, prices and money incomes rise. In foreign countries the opposite consequences ensue. Specie flows out, prices and money incomes fall. The reader will readily follow the further consequences. As money incomes rise in the United States, consumers will be led to spend more on imported goods. They will buy more of other imports (those which are not taxed) since these become cheaper as prices [144] abroad decline; and they will soon begin to buy somewhat more of the taxed goods themselves, as these show a decline from the enhanced price which appeared on the first imposition of the duty. On the other hand, consumers abroad will buy less of American articles than before. Their money incomes have been lowered, while they are faced by higher prices of the American goods. Exports from the United States will decline from the volume at which they stood at the outset. The movement of specie, and the consequent changes in prices and money incomes, will go on until the money value of the total imports again equals the money value of the total exports. Then new conditions of international trade will have been established; and these new conditions will become definitive; that is, will persist, Other things remaining the same, as long as the duties persist.

Eventually, then, equilibrium will be restored. American imports will be less, exports also less; the total volume of international trade will be diminished. But when the new equilibrium is reached, the terms of trade will have been altered to the advantage of the United States. Her people will have higher money incomes, and will be buying foreign goods which are lower in prices. Conversely, the people of foreign countries will have lower money incomes, and will be buying American goods which have risen in price. The Americans will be the gainers under the new terms of trade, the others the losers. For a given physical quantity of exports the Americans will be receiving a larger physical quantity of imports. The barter terms of trade will be changed to their advantage.

Needless to say, the tax which the American consumers pay, in the form of enhanced prices of the dutiable imports, is not to be regarded as a loss, not as something which offsets the gain to them from the better terms of trade. The proceeds of the tax serve to pay public expenses. Had it not been for them, some other levy would presumably have to be made. If it be suggested that the tax may lead to public extravagance, to expenditures that are wastefully made or bring no substantial gain to the community, it is sufficient to remark that this may happen under any and every [145] tax. The consideration of such possibilities is not germane to the matter in hand. A tax of this kind, or any particular tax, has in itself no tendency to promote misapplication of the public funds. It is the general state of a country's government that determines whether its tax revenue shall be well or ill used. So far as concerns the effects of international trade, it matters not what the government does with the proceeds of the tax; we may assume they are as well applied as the proceeds of other taxes.

Proceed now to a further set of possible consequences. So far it has been tacitly assumed that the duty is imposed for revenue purposes only. Suppose now that the duty is not merely for revenue, but is for protection. Its object then is to promote the production within the country of a part or the whole of the goods previously imported. In so far as it achieves this object, the results just indicated still ensue, and will be accentuated. Imports are cut down, not only because price rises and consumption becomes less, but because articles which before had been imported are made within the country. Specie flows into the country to a greater degree than under a revenue duty. Prices and money incomes rise more within its borders, and the barter terms of trade come to be altered even more to its advantage. Such imports as continue to come in—and, as will presently be explained, by no means all are likely to be shut out—will be procured at better terms.

In the case of protective duties, however, something more happens. There arises a real offset to the gain from better terms of trade. In so far as the taxed goods are produced within the country, there is loss, not gain. The price of these goods also is raised to consumers by the amount of the duty; that is, they are higher in price within the country than without. The enhancement of price, tho a tax in precisely the same sense as in the case of articles which continue to be imported, brings no revenue to the government. It goes to the domestic producers. And to them it represents no gain—that is, no special gain. It is in the nature of a bonus which makes it possible for them to conduct an industry in which the country has no sufficient advantage. [146] The returns to labor and capital in the newly stimulated industry are not higher than those current in the country at large. They may indeed be unusually large for a time after the first imposition of the duty; but if the industry be open to competition, they will come to the same level as elsewhere. That level, however, cannot be maintained unless prices of the goods are higher than they would be if imported; and to the community at large this difference in price represents pure loss.

In the case of a protective duty, then, there is a balancing of loss against gain; a loss which is overt and obvious, in the higher price of the goods whose domestic production is stimulated by the duty, and a gain, much less obvious, thru the more favorable terms of trade. There is no way of ascertaining which is the greater—whether the net result is positive or negative. Eventually the outcome would be affected on the one hand by the extent to which a disadvantageous domestic industry is brought into existence, on the other hand by those conditions of demand which determine the barter terms of trade in general.

In the exposition of this subject at the hands of the younger Mill, it was said that no gain at all accrues to a country from protective duties; these being believed to be "purely mischievous, both to the country imposing them and to those with whom it trades."⁰³⁴ This seems to be an error. If indeed all taxes on imports were protective, and if all were pushed so high as to attain unflinchingly the object of protection—domestic production of everything, all imports completely shut out—there would obviously be no gain from international trade, since the trade would cease once for all. But protection is never carried so far. As regards a particular article or group of articles, importation may indeed be entirely stopped. But other articles continue to come in, perhaps in large volume. There are many goods, such as the tropical products extensively used by the people of temperate zones, which it is so difficult to produce at home that an application of the protective policy to them is admittedly preposterous.

[147] They continue to come in free; and as regards them, the better terms of trade become more advantageous in consequence of the exclusion of the protected goods. And even as regards the latter, the same result may endure in part; that is, so far as the goods continue to be imported. The duty may be so nicely adjusted to the difference in money cost between domestic and foreign producers that domestic production is stimulated, while some imports nevertheless continue. A result of this kind is aimed at by persons who contend for a "competitive tariff"—one which shall leave a precise balance between domestic producers and their foreign competitors. A division between imported and domestic supplies also takes place, as has been elsewhere indicated, where the domestic industry is carried on under the conditions of diminishing returns.⁰³⁵ All in all, a rigorous and effective system of protection may yet permit a large volume of goods to come in from foreign countries. Those goods which continue to be imported are then obtained on the better terms of trade. There does exist this gain, to be reckoned as offsetting the direct loss caused by the protective duties.

A different ground for questioning whether in the end the attainment of any gain whatever will persist is that every country can play the same game. If the United States can get better barter terms of trade by imposing duties on goods coming from foreign countries, those other foreign countries can do the same by duties on goods coming from the United States. The application of the process on both sides not only increases the loss arising from the protective duties in themselves, but lessens the total gain from the division of labor that continues between the two sets of countries. True, some among them may perhaps retain a larger share of the remaining gain than others. But this preferential position, depending as it must on the elusive conditions of reciprocal demand, is neither easy to make sure nor easy to keep if once attained. Considering the trading world as a whole, and having in mind all the possibilities of retaliation, the quest of this sort of gain must be admitted to be highly hazardous. And if one finds the ordinary [148] arguments for protection untenable, nay intellectually repellent, and those for international cooperation and concord strong and appealing, there remains no inclination to commend this particular method of trying to capture a greater share of the total gain from trade between nations.

As a dry matter of analysis, however, it is to be said that, in this matter of import duties and the barter terms of trade, the position of the United States was long a comparatively advantageous one. The country's exports were chiefly foodstuffs and raw materials. Raw cotton bulked large among them; an article so much wanted by other countries that they never impose duties on it. Foodstuffs, again, were never subjected to duties by Great Britain, to which they went so largely; and the duties put on them by the countries of the Continent did not seriously check the imports that way. The countries to which the American exports mainly went were thus unwilling or unable to play the game of retaliation with much prospect of success. The protective duties imposed by the United States, however, did have the effect of checking some imports heavily and of stimulating domestic production to a corresponding extent; while yet other imports continued to come in. These other imports were in part goods of the protected class which came in over the barrier of the duties, such as wool, sugar, and sundry manufactured goods. In much larger part they were tropical or semi-tropical goods which were admitted free. It was the latter which came to predominate more and more largely in the import trade of the United States. As regards these, the barter terms of trade quite possibly become more advantageous, without any such offset as must be reckoned in respect of the protected goods.

So much as regards the general reasoning and some possibilities of its application. What bearing it may conceivably have on the actual course of the international trade of the United States will be considered in another connection.⁰³⁶