Tuesday, December 1, 2009

The Engine: Fractional-Reserve Banking (Part 1-Basic Overview)

Our banking system has been constructed around the principle of fractional reserves. There are pros and cons to this system---on the plus side, our economy is granted enhanced liquidity and the coordination of willing lenders (and inadvertent lenders, as we will see) and willing borrowers is greatly facilitated by the presence of an entity in the intermediary role, motivated by the opportunity to seek entrepreneurial profit. On the down side, it could be argued that almost all banks are technically insolvent and in violation of socially productive, ancient legal principles regarding the proper role of the custodian.

To discuss fractional-reserve banking, it is fruitful to begin with a few very simple examples. Let's pretend that a given individual, Depositor, has in his possession a prized Matisse painting. Concerned for the safety of the piece, he takes it to a secure warehouse facility owned and operated by another individual, Banker. Depositor and Banker enter into an agreement: in exchange for Banker providing security warehousing services for Depositor's Matisse, Depositor will pay Banker an agreed-upon amount. However, the agreement stipulates that Depositor can remove his painting from the warehouse at any time---the painting is thus available to Depositor on demand. Should Depositorank show up at the warehouse and demand his Matisse and Banker be unable to produce it, Banker will have broken the terms of the contract and a number of legal penalties may apply.

If Depositor were to entrust Banker with cash instead of a Matisse, it would certainly be appropriate for Banker to still charge a fee, probably 1-3% of assets, for performing security/warehousing as in the prior case, but also for bookkeeping and cashier functions. However, there would be some additional incentives for Banker if cash were deposited instead of the painting, because cash is considered a fungible good and a Matisse is not (two $10 bills are considered identical for purposes of legal tender; two paintings are not). The fungibility of money might tempt Banker to try to take Depositor's money for a personal investment, make a profit, and then be back in time to have it available for Depositor when he wants it. This fungibility argument, it turns out, was an important part of the legal defense of fractional-reserve banking and the claim that the bank is not duty-bound to perform a true custodian function with the depositor's money.

In our third example, a very different arrangement is constructed. Depositor brings cash to Banker and tells Banker that he does not need the money for a period of 1 year. He would like Banker to put that money into an investment that will generate 5% interest for Depositor over that investment horizon. Any interest over and above that 5% can be retained by Banker---this 5% represents the high-water mark and B has an incentive to find a good investment. Banker finds a third individual, Loan, who needs to borrow money to fund a business opportunity. Loan promises to pay Banker back, with 15% interest, in one year. Banker now takes Depositor's cash and gives it to Loan. In one year, Loan gives Banker the principal plus 15% interest, Banker gives Depositor his original money plus 5%, and Banker pockets the interest rate differential as his own profit.

Something very important about the deal: Depositor has no right to demand his money back on demand. In fact, he may need to accept that there is no guarantee that he will get his money back at all---the loan to Loan may or may not have adequate collateral or guarantees involved to cover the loss if Loan is unable to meet the terms of the deal. This all would need to be disclosed to Depositor, one way or another, but there is nothing intuitively unethical about Banker's behavior because Depositor offered his money to Banker expressly as risk capital and he knew that he would not have liquidity for the next year.

Our last example is where things get interesting, because it combines features of the first two in a way that is highly favorable to Banker. Depositor now deposits $1,000,000 in cash with Banker, intending for Banker to provide a safe haven for it (as in the warehousing Matisse case described in the first example). What follows will involve a few steps of basic T-account accounting, so bear with me. In the Anglo-Saxon accounting tradition, Banker debits the Asset side of his balance sheet with the $1,ooo,ooo as a "Cash" entry, and simultaneous credits his Liabilities with a matching entry of $1,000,000 under "Demand deposits" (reflecting that he is liable for $1,000,000, as if he was warehousing Depositor's Matisse, and Depositor can demand the money at any time).

Banker does not actually warehouse the money, however. He holds 10% back as a reserve and then proceeds to give $900,000 to Loan in the form of a loan. Banker then makes the following accounting entries: debit of $900,000 to "Loans" on the Asset side, credit of $900,000 to "Demand deposits" on the Liabilities side (under the Anglo-Saxon convention, a loan made is offset with the opening of a demand-deposit account in the borrower's name).

The bank now lists Assets as $1,900,000 (Cash + Loan), and Liabilities as $1,900,000 (two Demand deposits) . Something extraordinary has happened---the bank took $1,000,000 in real deposits and ended up with $1,900,000. In other words, $900,000 was created ex nihilo---from nothing---and put into the monetary system. In terms of actual "cash in the vault", however, the bank took a deposit of $1,000,000, retained 10% as a security reserve, and loaned the rest out to L in order to seek an entrepreneurial profit.

How far can the bank really go in terms of credit expansion? The process is theoretically governed by the formula x= d (1-c) / 1 + k (c-1), where:

x=maximum credit expansion
d=monetary value of deposits
c=reserve ratio (what the bank keeps as a reserve)
k=unused loans by borrowers

The lower the reserve ratio, c, and the higher the unused loan capacity of the bank, k, the greater the ability of the bank to expand credit by creating new ex nihilo money. If we assume that a banker's personal financial interests are directly correlated with the creation of ex nihilo money, as most of this money will be deployed as profit-generated loans for the bank, then we would predict that a banker will generally feel happiest when c is very low, ideally 1-2%, and k is very high, ideally closely to 100%. Laws may prohibit the banker from taking his reserve ratio beyond a certain downward limit, so k is often the target of innovations. As we will see a bit later, banks have learned to employ off-balance-sheet accounting entities called SPVs to remove loans from their balance sheets and open up new loan-origination capacity.

Welcome to the wonderful world of fractional-reserve banking. If this sounds dangerously close to a Ponzi scheme to you, then you probably have a good intuitive understanding of what is going on. The whole structure rests on the assumption that D will not want to withdraw a large percentage of his money at one time. If we use a more realistic example with many depositors and many loans, we could say that the structure rests on the assumption that any large withdrawal by a single depositor will be more than adequately covered by routine behavior from many other depositors. Under normal conditions, this is indeed a serviceable assumption. Under extreme conditions, depositors may find that they cannot remove their money from the bank because it is not holding adequate physical cash reserves---called "primary deposits"---to cover the demand deposits (in a 100% reserve system, the bank would be able to cover this scenario). The depositors have then become forced lenders and the true nature of the various relationships is brutally unveiled.

One may well ask how the fractional-reserve system was established to begin with. Like so many things in the economy, it emerged from a bootstrapped process of painful trial-and-error rather than a single catalytic event. An important aspect of its legal defense was that governments wanted to be able to access the credit that the fractional system made available.

There are some additional concerns related to fractional-reserve banking that involve the effects of excess credit creation and the punishment of saving at the expense of borrowing. However, our next installment will look at how we have collectively dealt with the problem of individual fractional-reserve banking failures, but at the risk of creating a larger probability of a more rare, but far more destructive, systemic banking crisis.

FURTHER READING: Huerta de Soto's "Money, Bank Credit, and Economic Cycles" provides an extraordinarily thorough account of the history of fractional-reserve banking, its legal standing, credit expansion, and so on. The following link has the Federal Reserve's spin on things: http://en.wikisource.org/wiki/Modern_Money_Mechanics .

No comments:

Post a Comment