Fractional Reserve Banking

Definition

Fractional reserve banking is the system in which commercial banks hold only a fraction of their deposits as reserves — vault cash plus balances at the central bank — and lend out the remainder. A bank that receives a $10,000 deposit does not store that money in a vault; it keeps perhaps $1,000 as a reserve buffer and lends the other $9,000. The borrower spends those funds, which land as deposits in another bank, which again keeps a fraction and lends the rest. Each cycle creates new deposits, multiplying the initial deposit into a larger money supply.

This mechanism is what allows a modern banking system to expand the money supply well beyond the base money created by the central bank. It is also what makes banks structurally vulnerable: they borrow short (deposits that can be withdrawn on demand) and lend long (mortgages, business loans), so a sudden surge in withdrawal demands — a bank run — can destabilize even a technically solvent institution.

Why it matters

Key takeaways

Banks create money through lending — when a bank extends a loan, it creates a new deposit, expanding the money supply. Loan repayment destroys that money.
The money multiplier (1 ÷ reserve ratio) is the theoretical upper bound on deposit expansion from a given base money injection. In practice, leakages (cash holdings, excess reserves) reduce the actual multiplier.
A bank run occurs when depositors simultaneously try to withdraw — the bank cannot honor all claims because most deposits have been lent out. Runs are self-fulfilling: rational even if the bank is solvent, because late withdrawers get nothing.
Deposit insurance (FDIC in the US) breaks the run dynamic by removing depositors' incentive to withdraw preemptively — if the government guarantees your deposit, you don't need to race to the teller.
The central bank acts as lender of last resort — lending to solvent but illiquid banks during a panic prevents runs from cascading into systemic collapse.
Capital requirements force banks to finance a fraction of assets with equity rather than deposits, absorbing losses before depositors are affected.

The bank run problem

Read it as: A bank run is a coordination failure — it is individually rational to withdraw early even if the bank is fundamentally sound, because late withdrawers lose everything. This makes the outcome self-fulfilling. Deposit insurance solves the coordination problem by making withdrawal timing irrelevant: depositors know they’ll be made whole regardless of when they act. The central bank’s lender-of-last-resort function handles the liquidity dimension — it can lend unlimited reserves to a solvent bank to meet any withdrawal surge.

How deposits multiply

The deposit expansion process

When the central bank injects $1,000 of base money into the banking system:

Bank A receives $1,000 in deposits, holds $100 (10% reserve), lends $900
The $900 lands in Bank B as a deposit. Bank B holds $90, lends $810
Bank C holds $81, lends $729 — and so on

The total deposit expansion converges to $1,000 ÷ 0.10 = $10,000 at a 10% reserve ratio. This is the theoretical money multiplier. The base money ($1,000) has been leveraged into a much larger money supply ($10,000) through the lending chain.

Why the multiplier is theoretical

The actual expansion falls short because: depositors hold some funds as cash (leakage from the banking system); banks voluntarily hold excess reserves as buffers (especially during uncertainty); and lending is constrained by borrower demand and creditworthiness, not just reserve availability. After the 2008 financial crisis, US banks accumulated trillions in excess reserves — base money did not translate into proportional money supply growth because lending demand was weak.