The Mystery Of Banking

by Murray N. Rothbard

  The point, however, is that there are good arguments either way on the money market fund, which highlights the grave problem the Fed and the Friedmanites have in zeroing in on one money supply figure for total control. Moreover, the money market fund shows how ingenious the market can be in developing new money instruments which can evade or make a mockery of reserve or other money supply regulations. The market is always more clever than government regulators.

  The Fed also issues an M-3 figure, which is simply M-2 plus various term loans, plus large denomination (over $100,000) time deposits. There seems to be little point to M-3, since its size has nothing to do with whether a deposit is a genuine time loan, and since term loans should not in any case be part of the money supply.

  The Fed also publishes an L figure, which is M-3 plus other liquid assets, including savings bonds, short-term Treasury bills, commercial paper, and acceptances. But none of the latter can be considered money. It is a grave error committed by many economists to fuzz the dividing line between money and other liquid assets. Money is the uniquely liquid asset because money is the final payment, the medium of exchange used in virtually all transactions to purchase goods or services. Other nonmonetary assets, no matter how liquid—and they have different degrees of liquidity—are simply goods to be sold for money. Hence, bills of exchange, Treasury bills, commercial paper, and so on, are in no sense money. By the same reasoning, stocks and bonds, which are mainly highly liquid, could also be called money.

  A more serious problem is provided by U.S. savings bonds, which are included by the Fed in L but not in M-2 or M-3. Savings bonds, in contrast to all other Treasury securities, are redeemable at any time by the Treasury. They should therefore be included in the money supply. A problem, however, is that they are redeemable not at par, but at a fixed discount, so that total savings bonds, to be accurately incorporated in the money supply would have to be corrected by the discount. Still more problems are proffered by another figure not even considered or collected by the Fed: life insurance cash surrender values. For money invested for policyholders by life insurance companies are redeemable at fixed discounts in cash. There is therefore an argument for including these figures in the money supply. But is the Fed then supposed to extend its regulatory grasp to insurance companies? The complications ramify.

  But the problems for the Fed, and for Friedmanite regulators, are not yet over. For should the Fed keep an eye on, and try to regulate or keep growing at some fixed rate, a raw M-1, or M-2 or whatever, or should it try to control the seasonally adjusted figure?

  In our view, the further one gets from the raw data the further one goes from reality, and therefore the more erroneous any concentration upon that figure. Seasonal adjustments in data are not as harmless as they seem, for seasonal patterns, even for such products as fruit and vegetables, are not set in concrete. Seasonal patterns change, and they change in unpredictable ways, and hence seasonal adjustments are likely to add extra distortions to the data.

  Let us see what some of these recent figures are like. For March 1982, the nonseasonally adjusted figure for M-1 was $439.7 billion. The figure for M-2 was $1,861.1 billion. If we deduct money market mutual funds we get $823 billion as our money supply figures for March 1982. There are at this writing no savings bonds figures for the month, but if we add the latest December 1981 data we obtain a money supply figure of $891.2 billion. If we use the seasonally adjusted data for March 1982, we arrive at $835.9 billion for the corrected M-2 figure (compared to $823.1 billion without seasonal adjustments) and $903.6 billion if we include seasonally adjusted savings bonds.

  How well the Reagan Fed has been doing depends on which of these Ms or their possible variations we wish to use. From March 1981 to March 1982, seasonally adjusted, M-1 increased at an annual rate of 5.5 percent, well within Friedmanite parameters, but the month-to-month figures were highly erratic, with M-1 from December 1981 to February 1982 rising at an annual rate of 8.7 percent. Seasonally adjusted M-2, however, rose at a whopping 9.6 percent rate for the year March 1981-March 1982.

  The numerous problems of new bank instruments and how to classify them, as well as the multifarious Ms, have led some economists, including some monetarists, to argue quite sensibly that the Fed should spend its time trying to control its own liabilities rather than worrying so much about the activities of the commercial banks. But again, more difficulties arise. Which of its own actions or liabilities should the Fed try to control? The Friedmanite favorite is the monetary base: Fed liabilities, which consist of Federal Reserve notes outstanding plus demand deposits of commercial banks at the Fed. It is true that Federal Reserve actions, such as purchasing U.S. government securities, or lending reserves to banks, determine the size of the monetary base, which, by the way, rose by the alarmingly large annual rate of 9.4 percent from mid-November 1981 to mid-April 1982. But the problem is that the monetary base is not a homogeneous figure: It contains two determinants (Federal Reserve notes outstanding + bank reserves) which can and do habitually move in opposite directions. Thus, if people decide to cash in a substantial chunk of their demand deposits, FRN in circulation will increase while bank reserves at the Fed will contract. Looking at the aggregate figure of the monetary base cloaks significant changes in the banking picture. For the monetary base may remain the same, but the contractionist impact on bank reserves will soon cause a multiple contraction in bank deposits and hence in the supply of money. And the converse happens when people deposit more cash into the commercial banks.

  A more important figure, therefore, would be total bank reserves, which now consist of Federal Reserve notes held by the banks as vault cash plus demand deposits at the Fed. Or, looked at another way, total reserves equal the monetary base minus FRN held by the nonbank public.

  But this does not end the confusion. For the Fed now adjusts both the monetary base and the total reserve figures by changes in reserve requirements, which are at the present changing slowly every year.

  Furthermore, if we compare the growth rates of the adjusted monetary base, adjusted reserves, and M-1, we see enormous variations among all three important figures. Thus, the Federal Reserve Bank of St. Louis has presented the following table of growth rates of selected monetary aggregates for various recent periods:7

  While total reserves is a vitally important figure, its determination is a blend of public and private action. The public affects total reserves by its demand for deposits or withdrawals of cash from the banks. The amount of Federal Reserve notes in the hands of the public is, then, completely determined by that public. Perhaps it is therefore best to concentrate on the one figure which is totally under the control of the Fed at all times, namely its own credit.

  Federal Reserve Credit is the loans and investments engaged in by the Fed itself, any increase of which tends to increase the monetary base and bank reserves by the same amount. Federal Reserve Credit may be defined as the assets of the Fed minus its gold stock, its assets in Treasury coin and foreign currencies, and the value of its premises and furniture.

  Total Fed assets on December 31, 1981 were $176.85 billion. Of this amount, if we deduct gold, foreign currency, Treasury cash and premises, we arrive at a Federal Reserve Credit figure of $152.78 billion. This total consists of:

  float-cash items due from banks which the Treasury has not yet bothered to collect: $10.64 billion

  loans to banks: $1.60 billion

  acceptances bought: $0.19 billion

  U.S. government securities: $140.4 billion

  Clearly, loans to banks, despite the publicity that the discount (or rediscount) rate receives, is a minor part of Federal Reserve Credit. Acceptances are even more negligible. It is evident that by far the largest item of Federal Reserve Credit, amounting to 79 percent of the total, is U.S. government securities. Next largest is the float of items that the Fed has so far failed to collect from the banks.

  Changes in Federal Reserve Credit may be shown by comparing the end of 1981 figures with the data
two years earlier, at the beginning of 1980. Total Reserve Credit, on the earlier date, was $134.7 billion, a rise of 13.4 percent in two years. Of the particular items, loans to banks were $1.2 billion in the earlier date, a rise of 33.3 percent in this minor item. The float’s earlier figure was $6.2 billion, a rise in this important item of 71.0 percent for the two years. The major figure of U.S. government securities had been $126.9 billion two years earlier, a rise of 10.6 percent in this total.

  If we take gold as the original and proper monetary standard, and wish to see how much inflationary pyramiding our Federal Reserve fractional reserve banking system has accomplished on top of that gold, we may note that the Fed’s total of gold certificates on December 31, 1981 was $11.15 billion. On this figure, the Fed has pyramided liabilities (Federal Reserve notes plus demand deposits at the Fed) of $162.74 billion, a pyramiding of 14.6:1 on top of gold. On top of that, however, the banking system had created a money supply totaling $444.8 billion of M-1 for that date, a pyramiding of 2.73:1 on top of the monetary base, or, an ultimate pyramiding of 38.9:1 on top of the Fed’s stock of gold.

  3. HOW TO RETURN TO SOUND MONEY

  Given this dismal monetary and banking situation, given a 39:1 pyramiding of checkable deposits and currency on top of gold, given a Fed unchecked and out of control, given a world of fiat moneys, how can we possibly return to a sound noninflationary market money? The objectives, after the discussion in this work, should be clear: (a) to return to a gold standard, a commodity standard unhampered by government intervention; (b) to abolish the Federal Reserve System and return to a system of free and competitive banking; (c) to separate the government from money; and (d) either to enforce 100 percent reserve banking on the commercial banks, or at least to arrive at a system where any bank, at the slightest hint of nonpayment of its demand liabilities, is forced quickly into bankruptcy and liquidation. While the outlawing of fractional reserve as fraud would be preferable if it could be enforced, the problems of enforcement, especially where banks can continually innovate in forms of credit, make free banking an attractive alternative. But how to achieve this system, and as rapidly as humanly possible?

  First, a gold standard must be a true gold standard; that is, the dollar must be redeemable on demand not only in gold bullion, but also in full-bodied gold coin, the metal in which the dollar is defined. There must be no provision for emergency suspensions of redeemability, for in that case everyone will know that the gold standard is phony, and that the Federal government and its central bank remain in charge. The currency will then still be a fiat paper currency with a gold veneer.

  But the crucial question remains: For there to be a gold standard the dollar must be defined as a unit of weight of gold, and what definition shall be chosen? Or, to put it in the more popular but erroneous form, at what price should gold be fixed in terms of dollars? The old definition of the dollar as 1/35 gold ounce is outdated and irrelevant to the current world; it has been violated too many times by government to be taken seriously now. Ludwig von Mises proposed, in the final edition of his Theory of Money and Credit, that the current market price be taken as the definition of gold weight. But this suggestion violates the spirit of his own analysis, which demonstrates that gold and the dollar are not truly separate commodities with a price in terms of the other, but rather simple definitions of unit of weight. But any initial definition is arbitrary, and we should therefore return to gold at the most conveniently defined weight. After a definition is chosen, however, it should be eternally fixed, and continue permanently in the same way as the defined unit of the meter, the gram, or the pound.

  Since we must adopt some definition of weight, I propose that the most convenient definition is one that will enable us, at one and the same time as returning to a gold standard, to denationalize gold and to abolish the Federal Reserve System.

  Even though, for the past few years, private American citizens have once again been allowed to own gold, the gold stolen from them in 1933 is still locked away in Fort Knox and other U.S. government depositories. I propose that, in order to separate the government totally from money, its hoard of gold must be denationalized; that is, returned to the people. What better way to denationalize gold than to take every aliquot dollar and redeem it concretely and directly in the form of gold? And since demand deposits are part of the money supply, why not also assure 100 percent reserve banking at the same time by disgorging the gold at Fort Knox to each individual and bank holder, directly redeeming each aliquot dollar of currency and demand deposits? In short, the new dollar price of gold (or the weight of the dollar), is to be defined so that there will be enough gold dollars to redeem every Federal Reserve note and demand deposit, one for one. And then, the Federal Reserve System is to liquidate itself by disgorging the actual gold in exchange for Federal Reserve notes, and by giving the banks enough gold to have 100 percent reserve of gold behind their demand deposits. After that point, each bank will have 100 percent reserve of gold, so that a law holding fractional reserve banking as fraud and enforcing 100 percent reserve would not entail any deflation or contraction of the money supply. The 100 percent provision may be enforced by the courts and/or by free banking and the glare of public opinion.

  Let us see how this plan would work. The Fed has gold (technically, a 100 percent reserve claim on gold at the Treasury) amounting to $11.15 billion, valued at the totally arbitrary price of $42.22 an ounce, as set by the Nixon administration in March 1973. So why keep the valuation at the absurd $42.22 an ounce? M-1, at the end of 1981, including Federal Reserve notes and checkable deposits, totaled $444.8 billion. Suppose that we set the price of gold as equal to $1,696 dollars an ounce. In other words that the dollar be defined as 1/1696 ounce. If that is done, the Fed’s gold certificate stock will immediately be valued at $444.8 billion.

  I propose, then, the following:

  1. That the dollar be defined as 1/1696 gold ounce.

  2. That the Fed take the gold out of Fort Knox and the other Treasury depositories, and that the gold then be used (a) to redeem outright all Federal Reserve Notes, and (b) to be given to the commercial banks, liquidating in return all their deposit accounts at the Fed.

  3. The Fed then be liquidated, and go out of existence.

  4. Each bank will now have gold equal to 100 percent of its demand deposits. Each bank’s capital will be written up by the same amount; its capital will now match its loans and investments. At last, each commercial bank’s loan operations will be separate from its demand deposits.

  5. That each bank be legally required, on the basis of the general law against fraud, to keep 100 percent of gold to its demand liabilities. These demand liabilities will now include bank notes as well as demand deposits. Once again, banks would be free, as they were before the Civil War, to issue bank notes, and much of the gold in the hands of the public after liquidation of Federal Reserve Notes would probably find its way back to the banks in exchange for bank notes backed 100 percent by gold, thus satisfying the public’s demand for a paper currency.

  6. That the FDIC be abolished, so that no government guarantee can stand behind bank inflation, or prevent the healthy gale of bank runs assuring that banks remain sound and noninflationary.

  7. That the U.S. Mint be abolished, and that the job of minting or melting down gold coins be turned over to privately competitive firms. There is no reason why the minting business cannot be free and competitive, and denationalizing the mint will insure against the debasement by official mints that have plagued the history of money.

  In this way, at virtually one stroke, and with no deflation of the money supply, the Fed would be abolished, the nation’s gold stock would be denationalized, and free banking be established, with each bank based on the sound bottom of 100 percent reserve in gold. Not only gold and the Mint would be denationalized, but the dollar too would be denationalized, and would take its place as a privately minted and noninflationary creation of private firms.8

  Our plan would at long
last separate money and banking from the State. Expansion of the money supply would be strictly limited to increases in the supply of gold, and there would no longer be any possibility of monetary deflation. Inflation would be virtually eliminated, and so therefore would inflationary expectations of the future. Interest rates would fall, while thrift, savings, and investment would be greatly stimulated. And the dread specter of the business cycle would be over and done with, once and for all.

  To clarify how the plan would affect the commercial banks, let us turn, once more, to a simplified T-account. Let us assume, for purposes of clarity, that the commercial banks' major liability is demand deposits, which, along with other checkable deposits, totaled $317 billion at the end of December 1981. Total bank reserves, either in Federal Reserve notes in the vaults or deposits at the Fed, were approximately $47 billion. Let us assume arbitrarily that bank capital was about $35 billion, and then we have the following aggregate balance sheet for commercial banks at the end of December 1981 (Figure 17.1).