Stabilizing the Dollar/Appendix 3
APPENDIX III
ALTERNATIVE PLANS
1. A Sound Alternative
A. Introduction. This book was written not so much in behalf of the specific plan described, which is regarded as the most practicable, as to show that the problem of stabilizing the dollar and the price level is soluble.
Many readers would like to know what alternative plans have been suggested. Of these the only one which seems worthy of careful consideration is that suggested, in a conversation with me, by Professor Gilbert N. Lewis, Professor of Chemistry of the University of California. He asked if it would not be possible to have paper certificates redeemable in the actual goods-dollars instead of in their gold equivalent.
It would, of course, be impracticable literally to maintain the "free coinage," i.e. deposit, of goods-dollars for certificates on the one hand and the free redemption of these certificates in goods-dollars on the other; because these goods-dollars would be too heavy, bulky, and perishable to use as reserve, as well as for other reasons. Nevertheless it would be entirely practicable to secure the desired regulation of the quantity and value of paper certificates by a simple device for indirect issue and redemption.
Such a system would first be launched by the conversion of our present gold certificates into certificates entitling the bearer to redemption in goods in accordance with the plan described below. All other money, bank notes, etc., would of course be redeemable in these goods-dollar certificates.
B. Redemption Warrants. The essence of redemption of these goods-dollar certificates is that their holder would be able, with certainty and without much trouble, expense, or delay, to exchange them for the commodities specified and in the quantities specified. This object could be substantially attained, as Professor Lewis suggested to me, by using the device of warrants for commodities as intermediate between money and commodities. This would break up redemption into two stages. The first stage would be when the holder of certificates would present them (in convenient lots of, say, one thousand dollars) at the Treasury and receive, in exchange, a set of separate specific warrants or orders, each warrant being for a specified amount of a particular one out of the collection of commodities represented. Thus one warrant might be for a thousand board feet of lumber, another for half a ton of sugar, etc., the entire collection constituting a thousand goods-dollars.
The second stage would be when these various warrants would be presented at separate offices for redemption in their respective commodities.
It is not necessary to discuss, at length, the exact organization of such offices. My object here is simply to show that, if we were willing to make the innovation of establishing such redemption offices, the plan would be economically feasible. Various methods could be used. Thus the Government could set up, in Washington, New York, or elsewhere, or in several different places, a great Government department store or agency which, whatever else it did, would receive these warrants and either hand over the goods from stock or agree, as immediately as practicable, to secure and deliver the goods called for. Another method would be for the Government to license a single private agency to conduct the business of redeeming the warrants for due consideration. A third method would be to license a number of such agencies, say one for each commodity. In this case, the most natural agencies would be existing large dealers in the various commodities concerned, a lumber merchant for redeeming lumber warrants, a wheat dealer for the wheat warrants, a coal dealer for coal, etc.
Nor do we need to discuss, in detail, the method by which the Government would reimburse any agencies it employed. It could pay lump sum retainers or the actual costs involved. The method simplest to understand would be for the coal dealer, for instance, after honoring a warrant for coal, to present a bill for said coal, at current prices, or at contract prices, to the Government, accompanying the bill with the warrant as evidence of the validity of the bill. This arrangement would be like that made for tourists by "Cook's" and other companies which sell warrants for meals and lodgings which are honored by hotels and later sent back to Cook's with bill for services rendered.
C. Unrestricted Redemption via Warrants. Whatever may be the business arrangements best suited for providing a working mechanism by which the Government would redeem certificates in the particular commodities which the certificates represent, our present interest lies in the working of that mechanism to stabilize the dollar.
The essence of the operations described is that certificate dollars are freely redeemable in goods-dollars (via warrants).
Such redemption would serve to correct any incipient depreciation of the certificate-dollar relative to the goods-dollar.
For, if the index number should rise much above par, i.e. if, in the open markets, the collection of goods constituting one thousand goods-dollars cost much more than a thousand dollars of certificates, recourse would be had to redemption. Speculators or warrant-brokers would arise who would find it profitable to gather together, say, $100,000 in certificates, redeem them in warrants, redeem those warrants in commodities, and sell these commodities in the open market for, say, $110,000 of certificates, thus making a profit of $10,000 (less expenses) on the series of transactions. As long as the index number were enough above par to make such operations pay, redemption would go on. The certificates so redeemed would be canceled, thus contracting the currency and reducing the index number toward par.
In short, if such redemption of money into commodities existed some people would refuse to patronize the markets at very high prices and, instead, patronize the Government which guaranteed to redeem certificates in goods.
D. Unrestricted Deposit of Goods-Dollars. So far we have considered only one of the two great regulators of the value of money; namely, unrestricted redemption of certificates in goods, constituting the outflow of money from circulation. The other is the "free coinage" or unrestricted deposit of goods, or some equivalent system of issuing certificates for goods, constituting the inflow of money into circulation. While, as already said, it would be impracticable to have literal composite goods-dollars brought to the mint to be exchanged for certificates exactly in the manner that gold is now exchanged, yet essentially the same result could be secured by the intermediation of warrants. The warrants would, in this case, pass from merchants to the Government instead of from the Government to merchants as in the operation of redemption.
To fix our ideas, we may suppose a licensed warrant-broker executing the following operations: First, with money (certificates) he buys up from miscellaneous sources, wherever he can get the lowest prices, the bill of miscellaneous goods constituting, say, $100,000 in goods-dollars. Some or all of these may be left in the custody of the respective dealers from whom he buys; but their ownership passes to him.
Secondly, he draws a sworn warrant for each of these lots of goods and presents at the Treasury the total assortment of such warrants, i.e. in the proportions required to constitute goods-dollars. He receives, in exchange, $100,000 of certificate-dollars. He has then virtually coined his goods into money or, at any rate, deposited 100,000 goods-dollars and received 100,000 certificate-dollars.
After the operation the Government then owns the miscellaneous bill of goods, or, let us rather say, a credit or right against the warrant-brokers to furnish 100,000 goods-dollars on demand or short notice (a right which would be enforced whenever the goods were needed for redemption of certificates).
The above process, simulating free coinage, would prevent the goods-dollar falling much below the certificate-dollar; for, as long as it is low enough to make such "coinage" of goods into money profitable, such coinage or deposit would continue. Thus if the merchant described above found that, at current prices, he could buy up the commodities constituting 100,000 goods-dollars for only $90,000 he would, after depositing them for $100,000, be making a profit of $10,000 (less expenses). The volume of money would then expand, prices would rise, and the profit on such operations would cease.
In short, the "free coinage" of goods-dollars would keep prices up because holders of goods, rather than sell at very low prices in the open market, would avail themselves of the Government's standing offer to buy 1000 goods-dollars for 1000 certificate-dollars.
E. Summary. These two processes, equivalent to our present free or unrestricted coinage and redemption, would keep prices from falling much below, or rising much above, par. They would thus put limits on the possible fluctuations of the index number, redemption taking place when the upper limit was reached and "coinage" or deposit taking place when the lower limit was reached. As long as the price level lay between these limits, there would be neither redemption nor "coinage."
It is scarcely worth while, as I am not advocating this plan, to go into much further detail. But it may be pointed out that, if desired, the limits to the index number may be narrowed if the Government would bear the expense in clerk hire, rent, interest, etc., involved in the broker's work of conducting his operations (just as, to-day, for an analogous reason the Government bears the expense of the Mint).
We have, so far, assumed that money, i.e. certificates, would come into being, as to-day, only by the act of "coinage," i.e. by the deposit of (warrants for) commodities and never by mere arbitrary issue to defray Government expenses, as in the case of "fiat money"; and, likewise, that money would pass out of existence only by the act of redemption, i.e. by the issue of warrants for commodities. The monetary system would then be strictly analogous to our present system, gold being replaced by a composite of commodities. It would not be a "fiat money" system.
2. The Same System Modified by the Omission of "Free Coinage"
We could, although with danger to the system, omit the "free coinage" feature, provided we permitted the issue of certificates for Government expenses and relied on such issue ceasing as soon as the resulting tendency toward redundancy brought about a demand for redemption. If, under such an arrangement, the Government should persist in overissuing paper certificates with one hand while redeeming them with the other, it would be losing through redemption what it would gain by the issue, in an "endless chain."
As to the opposite possibility, that of contraction, there would, if free coinage were not employed, be no safeguard. Nor would any be needed; for while, theoretically, the issue of certificates might be insufficient to keep the price level up to par, in actual practice the Treasury would be sure, in self-interest, to issue all it could without producing redundancy and loss from redemption.
The system described in this section would be exactly analogous to a system into which our present gold standard system would be transformed if we were to drop the free coinage, or deposit of gold (and permit, instead, the issue of certificates in payment of Government expenses limited merely by the obligation to redeem in gold).
3. The Same System Modified by the Omission of Redemption
To make our statement complete and symmetrical it should be added that we could imagine the opposite modification of the system, the "free coinage" feature being retained but the redemption feature omitted, the Treasury being allowed to issue certificates not only in exchange for (warrants for) commodities but also, at discretion, for expenses. But such a system would work only theoretically, i.e. on the assumption that the Treasury should systematically keep down its issues. It would be effectually stopped from undue contraction (were there any danger of that!) by the loss which would be imposed on it by warrant-brokers in demanding "coinage" of commodities. But, practically, the temptation would always be to expand and, as there would be no clear check on expansion, such a system would be almost sure to break down. It would be, in effect, what is called a "fiat money" system and little, if any, better than a pure "fiat money" system in which there is neither redemption nor coinage but only discretionary issue. Such a system is fundamentally unsound because there is nothing to check inflation. It would be analogous to a system into which our present gold standard system could be transformed by omitting redemption. Many writers (e.g. Parsons, Winn, and even Alfred Russel Wallace) have, it is true, seriously proposed such a discretionary system. But both experience and theory condemn it. No system yet proposed is really sound which omits the feature of redemption (nor is any system entirely sound if it omits the feature of deposit). Where paper money is vaguely assumed to "represent" commodities without any active redemption to make it good such representation is a mockery. Thus the famous, or infamous, "assignats" of the French Revolution were supposedly "based" on land but were in no way restricted thereby.
4. A Money Based on Labor
Others have suggested a plan somewhat analogous to the foregoing, the standard being a day's work of common labor.
We have as usual to consider the two fundamental operations of issue and redemption.
The plan provides for the virtual free "coinage" of such day's work into labor certificates by having the Government offer work on public roads or other public works issuing a fixed sum of money, e.g. three dollars for such day's work of common labor.
No provision for redemption is made, however. The certificates are receivable in taxes, but this does not make them convertible into day's work. The theory is that, should there be, at any time, an excess of certificates in circulation, their issue would be checked as workmen would refuse to work for the Government at the fixed price when, as a consequence of inflation, they could get more from private employers.
This system is in essence, therefore, the one-sided system described in § 3. It is as if we had free coinage or unrestricted deposit of gold for our present gold certificates without provision for redemption (although the certificates would, of course, be legal tender and receivable for taxes).
The faults of such a system are: (1) Lack of redemption as a decisive check on inflation. (2) The consequent temptation to inflate by issuing the certificates for general expenses. (3) The inconvenience and helplessness of the Government as to the amount of the public work it would thus give out. Sometimes workers would apply in large numbers and the Government would have to give them work, even if it did not really need them. At other times workers would apply in small numbers or not at all, because attracted, for the time, by private employers and the Government could not secure their services as it could not, without spoiling the currency, bid above its fixed price. Public works would thus be entirely subordinated to the maintenance of the currency. (4) The lack of definiteness of "a day's work of common labor" and the lack of its fluidity.
The question of the relative virtues of the labor standard and the commodity standard is discussed in my Purchasing Power of Money, Chapter X, § 4.
5. Govermnental Control of Gold Production
Mr. B. M. Anderson, Jr., suggests international Governmental control of gold mining, or a variable tax on gold mining. The former has already been mentioned.[1] The latter would be unjust to gold miners and, for that reason alone, impracticable. The plan proposed in this book must not be confused with such a plan. It is not a plan to control the output of gold. As shown in Appendix II, § 4, the gold miner would not be adversely affected but would share in the general advantage and prosperity which the plan would bring.
6. The Tabular Standard
As is shown under "anticipations" in Appendix VI, § 3, D, the idea of a tabular standard is a very old one, and, as shown under "precedents," in Appendix V, it has in a number of instances, notably during the Great War, been actually employed.
The proposal is, not to change the monetary standard itself but to correct its injustices in any contract by supplementary payments from the debtor to the creditor or vice versa according to an index number. If a debt for $1000 were contracted in 1900 and paid in 1920 and if the index number in 1920 were 250, on the basis of 100 for 1900 the debtor who had engaged to pay by the tabular standard would, in 1920, owe 250% X $1000, or $2500; that is, he would supplement the $1000 which his debt calls for by $1500 under the tabular standard agreement.
This plan would, apparently, accomplish everything which the plan proposed in this book would accomplish and without disturbing our monetary system in the least.
Practically, however, it would never accomplish more than a small fraction of what a true stabilization of the dollar would accomplish and, if widely used, would really cause considerable disturbance, of one sort and another.
As a makeshift in an emergency this plan is worth while, especially for correcting wages, but its inconveniences stand in the way of a wide adoption, especially in ordinary times. In the absence of a real standardization I favor[2] it most heartily and hope that it may serve as a stepping stone toward something better.
The two chief objections are (1) the inconvenience of calculating (which would be like that which would be caused if we were to use as our yardstick of length the height of a barometer and had to employ a new correction factor each day for selling cloth) and (2) the trouble which would come from the fact that the tabular standard would only be partially employed. Thus if a merchant corrects the items only on one side of his ledger by an index number, his profits would be destabilized rather than stabilized.[3]
- ↑ See Appendix I, § 1, L.
- ↑ See Irving Fisher, "Adjusting Wages to the Cost of Living," Monthly Labor Review, November, 1918.
- ↑ See Irving Fisher, The Purchasing Power of Money, New York (Macmillan), p. 336, and "Rejoinder by Professor Fisher," American Economic Review, June, 1919, pp. 256-262.