The essence of modern banking is arbitrage: banks borrow from a group of people who are in a position to supply funds (depositors) and lend to another but not completely disjoint group (borrowers) who demand funds. In order to compensate depositors for the use of their funds, banks pay them interest; and to compensate themselves for the risk that inheres in lending, banks charge interest on the funds lent to borrowers. To the extent that they arbitrage successfully, i.e., receive more interest from borrowers than is paid to depositors, and receive principal from borrowers and can return it to depositors, banking is a profitable exercise (see in particular Vera C. Smith, The Rationale of Central Banking and the Free Banking Alternative, The Liberty Fund, 1936, 1990, ISBN: 0865970866; Murray Rothbard, The Mystery of Banking and Murray Rothbard, A History of Money and Banking in the United States: The Colonial Era to World War II, Ludwig Von Mises Institute, 2002, ISBN: 0945466331).

A critical ingredient of banks’ success and failure is the creation and destruction of demand deposits. A modern bank lends what would otherwise be idle balances from depositors’ accounts by creating (or adding to) borrowers’ demand deposits. Borrowers then write cheques in order to spend the borrowed funds. Demand deposits facilitate commercial transactions and are an integral part of the money supply. A depositor’s signature on a cheque authorises his bank to pay a portion of his demand deposit to the bearer of the cheque. As Gene Smiley (Rethinking the Great Depression: A New View of Its Causes and Consequences, Ivan R Dee, 2002, ISBN: 1566634725) notes, there is a crucial difference between demand deposits and other types of money.

By the late nineteenth century, banks in America, Australia, Britain and other countries had generally become “fractional reserve” banks. When a bank clears a cheque it debits a demand deposit; and when a bank’s customer deposits funds into an account the bank credits the customer’s demand deposit. Most of the time, the amount of currency paid when cheques are cleared approximates (and thus offsets) the amount received in the form of new deposits. Given that it is safer and more convenient to hold one’s liquid funds in the form of a demand deposit in a bank rather than notes and coins stuffed under the mattress or buried in the back garden, under normal conditions it is unlikely that all or even many of a bank’s depositors will simultaneously seek to convert their demand deposits into currency. Hence the development of “fractional reserve” banking.

Fractional reserve banks retain only a fraction of their deposit liabilities in their vaults (i.e., in the form of currency reserves); instead, most of their deposit liabilities are lent to borrowers. Banks lend in order to generate the income required to pay the bank’s staff and other expenses, provide dividends for their shareholders and pay interest to their depositors. Other things equal (and assuming that the bank arbitrages successfully between depositors and borrowers and that many depositors will not simultaneously seek to convert their demand deposits into currency), the lower the fraction of deposits held in reserve the more profitable the bank. Hence the perennial risk that banks lend too aggressively (see in particular James Grant, Money of the Mind: Borrowing and Lending in America from the Civil War to Michael Milken, Noonday Press repr. ed. 1994, ISBN: 0374524017).

A Stylised Example

Gene Smiley provides a stylised example that shows how fractional reserve banking can create (and destroy) money. Let us assume that a gold miner extracts gold nuggets that are worth $5,000, and that a local mint transforms them into $5,000 of gold coins. Also assume that the miner deposits the coins (i.e., converts the coins into a demand deposit with a balance of $5,000) at his local bank. For convenience let us refer to this bank as Bank A. Further, let us assume that the law requires that banks hold a minimum percentage of their deposits in cash reserves composed of either gold coins or currency (bank notes). More specifically, assume that banks are required to hold 12.5% of their deposits as reserves; that as a matter of practice banks retain some additional margin of reserves (above the required minimum) as a precaution against unexpected deposit withdrawals; and that therefore banks normally hold 15% of their deposits in gold and currency.

This amount sits either in the bank’s vault or is a demand deposit at another (probably larger) bank. Accordingly, Bank A would retain $750 of the miner’s $5,000 deposit in reserves and would seek to lend the remainder ($4,250) to creditworthy borrowers. If a borrower obtained a loan, then the bank would lend the money by creating a new demand deposit in the borrower’s name (or by adding to the borrower’s current demand deposit) of an amount equal to the loan. Once the bank had lent its new excess reserves of $4,250, it could make no new additional loans until it obtained additional excess reserves. As Smiley notes, however, the story does not end here.

For simplicity let us assume that a single manufacturing firm borrowed the entire $4,250 and that it used this amount to construct additional space at its factory. Assume as well that the firm that extended the factory received the cheque for $4,250 and that it deposited the cheque in its bank (which we will call Bank B). Bank B sends the cheque to Bank A, where $4,250 is deducted from the borrowing firm’s demand deposit and $4,250 (either in the form of currency, or, more likely a bank cheque) is sent to Bank B. Accordingly, Bank A no longer has excess reserves from the miner’s deposit of $5,000. But Bank B has. More precisely, on its balance sheet it now has a new liability (demand deposit) of $4,250 and new asset (additional cash reserves) of $4,250; but given that it holds 15% of deposits as reserves it will retain only $637.50 in reserves against this new deposit. Bank B can lend its new excess reserves of $3,612.50 by creating (or adding to) a demand deposit for a borrower. When a second borrower spends the $3,612.50 and his cheque clears Bank B, this bank (like Bank A) would have no excess reserves available for loans. But Bank C has. Let us say that the $3,612.50 cheque is deposited in a demand deposit in another bank, and that this bank is called Bank C. It now has excess reserves of $3,070.63 and can lend this amount by creating (or adding to) a borrower’s demand deposit.

Consider the process thus far. A miner has mined $5,000 worth of gold and, after having it struck into coins, deposits it in Bank A. The new $5,000 of gold has been converted into a new demand deposit of $5,000 and the money supply has increased by $5,000. Because banks keep only a fraction of their deposits in reserves, Bank A lent the excess reserves and those became a new demand deposit of $4,250 at Bank B. Bank B then had excess reserves, which became a new demand deposit of $3,612.50 at Bank C. In a fractional-reserve banking system, the “hard money” of $5,000 (i.e., the gold the miner found) has set in train a process that has increased demand deposit money by $12,862.50. Further, the process is not finished because Bank C now has excess reserves. This process of expanding the money supply through the creation of demand deposits can continue until all of the $5,000 in gold is transformed into required and precautionary reserves. Given our assumptions, the miner’s production of $5,000 of money could potentially increase the money supply by $33,333.33 (i.e., $5,000 ÷ 0.15). Fractional-reserve banking thus tends inherently to change the money supply; further, the smaller the reserve fraction the greater the resultant change of the money supply. Given this leverage, relatively small changes in reserves typically create much larger changes in the supply of money.

Fractional Reserve Banking and Monetary Disturbance

Smiley cites an historical example to illustrate the problems that may arise – indeed, inhere – in fractional reserve banking. On 14 October 1907, an unusually large number of the depositors of five banks in New York City began to convert their demand deposits into currency. The depositors’ preference for currency over deposits exceeded the banks’ ability to convert deposits into currency: trouble at these institutions therefore brewed. On 21 October, the Knickerbocker Trust Co. experienced a severe “run” on its deposits and failed; and on 24 October, depositors at the second largest trust company in the City began a run on their deposits.

Chastened by this unexpected development, increasingly worried depositors at other banks began to convert their deposits into currency; in a chain reaction, increasingly concerned banks outside New York that held deposits with banks in the City now began to convert those deposits into currency and gold; and depositors at banks in other Eastern cities, observing the unfolding panic in New York, began to convert the deposits at their banks into currency. By late October 1907 a nation-wide banking panic had developed.

Under a fractional reserve banking system, banks cannot instantly pay cash to large numbers of their depositors because the bulk of the deposits are held in an illiquid form, i.e., loans and income-earning securities rather than currency and gold. And in the absence of a central or reserve bank, commercial banks can turn only to the strongest and most solvent among their ranks, if such a bank exists, in order quickly to obtain cash. If large numbers of banks sell their loans and securities in order to obtain cash for their panicky depositors, their concerted action would drive down the prices of the loan books and securities. Under these circumstances the banks would not have sufficient assets to cover their liabilities and would thus be bankrupt.

In 1907 J.P. Morgan’s rock-solid balance sheet and the confidence he and it inspired saved the day and eventually halted the panic. Banks were able to stem the redemption of deposits into currency and gold. They continued to clear cheques and borrowers continued to repay their loans. Just as they had ended previous banking panics, these actions ended the Panic of 1907 (as economic historians subsequently dubbed it). But in its immediate aftermath, as a safety precaution banks increased the percentage of their deposits they kept as cash reserves. And depositors, for a while, held more of their capital in currency and less in demand deposits.

To rebuild their reserves – for example, to increase the percentage of deposits held as reserves from 15% to 20% – banks must reduce their net lending. If borrowers repaid $100,000 of their outstanding loans, for example, then banks might create only $85,000 of new loans. They could do this by increasing interest rates or the terms and conditions of collateral and repayment, or by “calling” (i.e., demanding immediate repayment of) loans: either would reduce the demand for loans. Under these conditions the demand deposits “destroyed” by the repayment of old loans would be larger than the deposits “created” by new loans. Increasing the fractional reserve ratio thus tends to decelerate, halt or reverse the growth of the money supply. The miner’s new $5,000 in gold would result in $25,000.00 of new demand deposits at a 20% reserve ratio versus $33,333.33 at a 15% ratio. Further, and perhaps more importantly, at a 20% ratio each dollar that a depositor converted into currency rather than hold as a demand deposit would require that the banking system destroy $5 of demand deposits. At a 15% reserve ratio each dollar of demand deposits converted into cash required the destruction of $6.67 of demand deposits.

Under a regime of fractional reserve banking, then, relatively small changes in reserves eventually generate much larger changes in the money supply. In 1907 banks responded to the October runs by accumulating but not lending excess reserves. The money supply subsequently shrank, the pace of economic activity (which had hitherto depended upon debt finance) decelerated and prices (including the price of labour) fell precipitously. The rapid decrease of many prices, which was unrelated to the preferences and demands of consumers, drastically but temporarily weakened the ability of the price mechanism smoothly, accurately and efficiently to co-ordinate the actions of capitalists, producers and consumers. Only when the structure of production adjusted to these new and chastened conditions could prices transmit accurate signals, growth resume and prosperity return.

In 1907, then, a monetary disturbance and the institution of fractional reserve banking transformed what might otherwise have been a brief and relatively minor contraction into a rapid and severe decline of economic activity that persisted until June 1908. During the 1920s and early 1930s, much more severe, complex and extended monetary disturbances occurred. These disturbances, together with politicians’ and policymakers’ utter unwillingness to let the price mechanism operate freely, transformed what might have been a short and sharp contraction of a 1907 or 1920-21 type into a depression of unprecedented severity that lasted throughout the decade.

...continued in Part IV