which, in the long run, he suffers; it is changes in the rate of inflation or of contraction that produce marked and damaging effects at the critical points of business life. How does a slow and steady appreciation of any metallic currency affect the relations of business classes? Does it rob borrowers and enrich lenders? Does it favor the consumers

by giving low prices, and hurt producers in the same degree? Does it tax enterprise and paralyze the nerves of business? The answer is an emphatic No. Steadiness in the rate of appreciation of money is the salvation of business. Not by one iota can such a slow and steady movement, in itself alone, rob the borrowing class. This is a sweeping claim; let us examine it.

It is coming to be generally known that true interest is governed by the "marginal productivity" of capital. As the utility of the final increment of a commodity fixes the price that a seller can get for his whole supply, so the productive power of the final unit of capital expresses what the owner of capital can get by lending his entire supply. This earning capacity expresses itself in a percentage of the capital itself. If the final unit can create a twentieth of itself in a year, any unit can get for its owner about that amount.

In assuming that capital earns a twentieth of itself in a year, we may use a commodity standard of measurement. A grocer's capital of twenty barrels of sugar may become twenty-one barrels, and his flour and his tea may increase in a like proportion.1 In the simplest illustration that could be given of a capital earning five per cent a year, we should assume that each kind of productive instrument in a man's possession increases in quantity, during the year, by that amount. If he be a manufacturer, his mill becomes a hundred and five feet long,

1 I cannot here express in any completeness the "final productivity" theory of interest. It is safe to assume that it is winning acceptance. One feature of this law of interest is that the productive power of all capital tends to equal that of the marginal social unit of it. In the illustration I assume that all capital creates for the man who uses it a twentieth of itself in a year, and that this increase is capable of being measured in kind. A more complete statement would fully describe the unit of measurement used.

instead of a hundred feet. It contains twenty-one sets of woolen machinery, instead of twenty. The flow of water that furnishes power becomes by five per cent more copious; and the stock of goods, raw, unfinished and finished, becomes larger by the same amount.

Of course such a symmetrical enlargement of all kinds of goods could never actually take place, for some things increase in quantity more than others. The illustration shows, however, what fixes the rate of interest: it is the self-increasing power of a miscellany of real capital. If the mill, the machinery, the stock, grow in quantity at the five per cent rate, that is the natural rate of interest on loans of real capital. The lender gives to the borrower twenty units of "commodity" and gets back twenty-one. If marginal social capital, consisting of commodity and measured in some way in units of kind, has the power to add to itself in a year one unit for every twenty, lenders will claim about that amount, and borrowers will pay it.

How does the real earning capacity of capital in concrete forms reveal itself? How does the grocer know that he can make five per cent with the final unit of capital that he borrows? Not by the fact that each lot of twenty barrels of sugar gains one barrel, that each lot of twenty pounds of tea gains one pound, and so on. If there were to be such a symmetrical all-around increase in the commodities in the man's possession, his shelves, counters, bins, tanks, would have to enlarge themselves in the same ratio. In the case of a manufacturer the mill would have to elongate itself by one foot for every twenty, as in the foregoing illustration, and the machinery and all the stock would have to grow in the same ratio. The land and the water power would have to enlarge themselves by the same constant fraction.

I have said that such a thing does not take place. The general amount of capital goods of every kind enlarges; but the enlargement is in practice computed in monetary value, and in no other way. The whole outfit becomes worth more than it was. The increase in monetary value gauges the claims of the capitalist. If the stock of goods has grown generally larger, and

if prices have fallen, the claim of the capitalist will fall short of equalling the actual increase of the merchandise.

The increase in goods of different kinds is of course unsymmetrical. If the man is a manufacturer, his mill and his water power have probably not increased. He may have some more machinery, and he has more raw materials and more goods, finished or unfinished, than he had when he took his last inventory. If he has not more goods of these kinds, he has something that represents them; and the effect on his fortunes is as if the mill had stretched itself, and as if the machines and other capital had multiplied, all in the same ratio.

The man figures his gains in real wealth by the use of money. At the end of the year he makes a list of all his goods, attaches prices to them, and sees what the value of the stock has become by the year's business. He compares the total value in money of the goods on hand in January, 1896, with that of the stock of January, 1895. If he has bought and sold for cash only, and if during the year he has drawn for his maintenance only what he has earned by labor, the excess of value on hand at the beginning of the year 1896 informs him what his capital has earned during the preceding twelve months.

If prices have remained stable, the earnings of the capital as expressed in money will accurately correspond with the earnings as computed in commodity. It is as if the five per cent increase of the sugar and the flour of our first illustration, or of the mill and the machinery of the second, had taken place. It could then, by a sale, be converted into a five per cent increase in money. By selling the stock at its market value the merchant could realize five per cent more than the original stock cost him.

If money has gained one per cent in its purchasing power, or if prices at the end of the year are by so much lower, the inventory will show, in terms of money, only a four per cent gain. Now the real increase of concrete capital is still five per cent, and that, by the law of interest, is what the capitalist can claim in commodities. This claim is met by an actual payment in money of four per cent. Give to the capitalist, in Jan

uary, 1896, a dollar and four cents for every dollar that he loaned in January, 1895, and you enable him to command a hundred and five units of commodity for every one hundred that he commanded at the earlier date. You give him by a reduced monetary payment what is equivalent to the real increase of capital.

It is the increase of capital in kind that fixes the rate of loan interest. I have been careful not to claim for this part of the adjustment any unerring accuracy; for the marginal-productivity law does not work without friction. With real capital creating five and a half per cent, the lender might get only five. When, however, the play of forces that fixes real interest has had its way and has determined that, in commodity, capital shall secure for its owners five per cent a year, that amount is unerringly conveyed to them by the monetary payments that follow. If, by paying four per cent as interest, the merchant, in the illustrative case, makes over to the lender of capital that part of the increase of goods that by the law of interest falls to him, four per cent is the rate that the loan in money will bring. This is on the supposition that the change in the purchasing power of money is perfectly steady. If it is unsteady, effects will follow that are of much consequence.

In a late number of this QUARTERLY,1 I expressed the opinion that changes in the purchasing power of a currency produce an effect on the rate of interest on loans of "money." If, with a currency of perfectly stable value, the interest on loans is five per cent, corresponding to the earnings of real capital, then a gain in the purchasing power of the currency of one per cent a year has the effect of reducing nominal interest to four per cent. The debtor then really pays and the creditor really gets the same percentage as before of the actual capital loaned. The borrower, the entrepreneur in the case, finds at the end of the year that he has more commodities by five one-hundredths than he had. He must pay the equivalent of this to the lender.2 With money of stable purchasing power it takes five new dollars for every hundred to do it; but with money that 1 Vol. x, p. 389 (September, 1895).

2 Of course the gains of different entrepreneurs vary. It is the earnings of marginal social capital that really determines interest.

gains in its power to buy goods at the rate of one per cent a year it takes only four.

Since this view was submitted, an acute critic1 has expressed the opinion that the rate of interest on loans is, indeed, varied because of the appreciation of the coin, and that the variation is in the direction indicated; but he has claimed that the variation is not as great as the appreciation of the coin. He has cited statistics to prove that gold has recently gained in purchasing power, that the rate of loan interest has been reduced because of this gain, and that the amount of the reduction is only about a third as great as the amount of the appreciation of gold. Where the gain in the purchasing power of money would theoretically call for a certain lowering of the nominal rate of interest, the facts cited appear to show that interest has been lowered, but by a smaller amount.

Now without meeting this contention with a tant pis pour les faits, I propose to reassert the former doctrine, and to claim that the rate of interest on loans is, in the long run, reduced by an amount that accurately corresponds with the appreciation of the monetary metal wherever the appreciation is steady. This law, as I believe, works with a precision that is unusual in the case of economic laws; and I have seen no facts cited which do not seem to me to be consistent with this view. Loan interest varies more or less from the marginal earnings of capital. Interest as paid in money accurately expresses interest as determined in kind by the play of economic forces.

It is possible that, where this subject has been considered, the impression may prevail that this reduction in the nominal rate of interest is the result of foresight on the part of borrower and lender. According to that view, both parties look forward to the time when the loan will be paid. The borrower sees that, although by means of his business he may have at the end of a year five per cent more of commodity in his possession, prices. will probably have fallen so as to enable him to realize in money only four per cent. On the other hand, the creditor will see that

1 The criticism referred to was made by Professor Irving Fisher of Yale University, in a paper read before the American Economic Association, at the annual meeting lately held at Indianapolis.