Fractional-reserve banking is the system, used in nearly every modern economy, in which banks hold only a fraction of their customers' deposits as cash or reserves at the central bank and use the remainder to make loans or buy securities. Because a loan is paid out as a new deposit, the process expands the money supply beyond the amount of base money issued by the central bank.
The system relies on the fact that depositors rarely all withdraw at once. It lets banks perform maturity transformation, funding long-term loans with short-term deposits, but it also exposes them to runs. Its alternatives are full-reserve banking and narrow banking, and it is the model that stablecoin laws deliberately do not allow for issuers.
History
The practice is usually traced to goldsmiths in seventeenth-century London, who issued receipts for gold left in their vaults and found they could lend part of the gold because only a share of receipts was presented at any time. Banks of issue later did the same with banknotes and then with deposit accounts. Repeated panics in the nineteenth and early twentieth centuries led to central banks acting as lenders of last resort, to the creation of the Federal Reserve System in 1913 and, after the bank failures of the early 1930s, to federal deposit insurance in the United States.
How money creation works
The textbook account uses the money multiplier. If banks must hold a reserve ratio , the maximum amount of deposits that can be supported by a given amount of reserves is:
Here is the maximum level of deposits, is total reserves and is the reserve ratio written as a decimal, so the multiplier is . With a reserve ratio of 10% and reserves of 1,000, deposits could in theory reach : the original 1,000 deposit plus 9,000 of new money created as loans are re-deposited and re-lent.
Central banks now describe the process differently. A Bank of England paper published in 2014 stressed that banks create deposits when they decide to lend, limited mainly by profitability, capital rules and the interest rate set by the central bank, and obtain reserves afterwards. On that view the multiplier is a poor guide to the link between the monetary base and broad money, which varies widely over time.
Reserve requirements and other safeguards
A reserve requirement is the legal minimum share of deposits a bank must hold in reserves. Many central banks have reduced or abandoned them. The Federal Reserve cut its requirement to zero in March 2020, and the euro area ratio has been 1% since 2012, while some emerging-market central banks still use changes in reserve ratios as a policy tool. In practice, banks' safety now rests on other tools:
- Capital requirements. Basel rules require banks to fund part of their assets with equity that can absorb losses.
- Liquidity rules. The liquidity coverage ratio requires enough high-quality liquid assets to survive a 30-day stress.
- Deposit insurance. Schemes such as the FDIC guarantee deposits up to a limit, reducing the incentive to run.
- Central bank lending. Discount window and emergency facilities lend to solvent banks against collateral.
Bank runs and criticism
Because a bank cannot repay all deposits at once, a loss of confidence can become self-fulfilling: a bank run. Northern Rock in 2007 and Silicon Valley Bank in 2023 showed that runs still happen, now at the speed of mobile banking. Critics from different schools, including the authors of the 1930s Chicago Plan and later proponents of sovereign money, argue that money creation should be separated from lending. Supporters reply that fractional reserves allow credit to flow to households and firms at lower cost, and that liquidity risk can be managed by regulation.
Fractional reserves and stablecoins
Stablecoin regulation is built around the opposite principle. The US GENIUS Act requires a payment stablecoin to be backed at least one to one by cash, deposits, short-term Treasury bills and similar assets, and bars issuers from lending out reserves. The EU's MiCA requires full backing of e-money tokens and asset-referenced tokens. The aim is that a stablecoin behaves like a fully backed claim rather than a bank deposit, which is why tokenised deposits, issued by banks, are treated as a separate category.
External links
Ref 1030000931