What You Think Banks Do, and What They Actually Do
The common understanding of banking goes something like this: you deposit $1,000. The bank keeps some as a reserve, lends out the rest, earns interest, and pays you a fraction of that interest as a reward for letting them use your money. The bank is an intermediary, connecting savers to borrowers.
This understanding is not entirely wrong. But it omits a critical detail that transforms the entire picture: banks do not merely lend out existing deposits. They create new money when they make loans.
This is not a conspiracy theory or a fringe economic position. It is standard monetary economics, confirmed by central banks themselves. The Bank of England published a paper in 2014, Money Creation in the Modern Economy, stating plainly:
"Whenever a bank makes a loan, it simultaneously creates a matching deposit in the borrower's bank account, thereby creating new money."
Understanding how this works, its history, its mechanics, and its consequences, is essential context for understanding inflation, economic cycles, financial crises, and why Bitcoin's design represents a fundamentally different approach to money.
From Goldsmiths to Banks: The Origins of Fractional Reserves
The story begins in medieval Europe, long before central banks or printed currency. Gold and silver coins were the primary medium of exchange, functional and scarce, but inconvenient to carry in quantity. Merchants transporting substantial sums risked robbery on every journey.
Goldsmiths, who already maintained secure vaults for their trade, began offering storage services to wealthy merchants. You would deposit your gold, receive a paper receipt, and retrieve the gold on demand. The receipts themselves proved useful: if the goldsmith's trustworthiness was well-established, you could simply transfer the receipt rather than physically move the gold. Over time, these receipts began circulating as money. The paper stood in for the metal.
Goldsmiths noticed something significant: most of the time, only a small fraction of depositors came to claim their gold on any given day. The metal sat in the vault. This created an opportunity, and a temptation.
If 100 units of gold were deposited and only 10 were typically needed at any moment, a goldsmith could issue receipts for 150 units and lend out the extra receipts at interest. As long as redemption claims remained within the actual gold on hand, no one would notice. The goldsmith was earning interest on money that had no metal behind it.
This is fractional reserve banking in its original form. The "fraction" refers to the fraction of deposits actually held as reserves; the rest is lent out, or, in the modern formulation, created as new deposits.
The Bank of England and Institutionalized Money Creation
The Bank of England, founded in 1694, formalized and scaled this practice at a national level for the first time. It was established to lend money to the English Crown, primarily to finance the Nine Years' War, and issued banknotes backed by government debt rather than fully by gold.
The bank pioneered what economists call the money multiplier: the mechanism by which a single unit of base money can support a multiple of that amount in total bank deposits. By accepting deposits, making loans, and accepting the resulting deposits again, banks collectively expand the money supply far beyond the base that any central authority directly controls.
Over the following two centuries, this model spread across Europe and the Americas, becoming the universal foundation of commercial banking. Every major financial system today operates on fractional reserve principles.
How It Works Today: The Mechanics of Money Creation
In the modern banking system, money creation works somewhat differently from the goldsmith model, but the essential mechanism is the same.
Here is a concrete example of how a new loan creates money:
- You apply for a $300,000 mortgage to buy a house.
- The bank's credit officers approve the loan.
- The bank does not locate $300,000 in existing deposits and transfer them to you. Instead, it opens a new deposit account in your name and credits it with $300,000: money that did not previously exist in any form.
- You use that deposit to pay the home seller; the funds move to the seller's account.
- The seller's bank now holds new deposits, a portion of which it can use to support new loans of its own.
- The cycle continues: each new deposit becomes the raw material for additional loans and deposits.
This is why economists distinguish between the monetary base (notes and coins in circulation plus bank reserves held at the central bank, also called M0) and the broader money supply (M1, M2, M3), which includes all bank deposits:
- M0 / Base Money: Physical currency plus central bank reserves, directly created by the central bank
- M1: M0 plus demand deposits (checking accounts), the most liquid money in everyday use
- M2: M1 plus savings accounts, money market funds, and small time deposits
- M3: M2 plus large time deposits, institutional funds, and other near-money instruments
In most developed economies, M2 is four to ten times larger than M0. The difference represents money created by the banking system itself through lending, not by the central bank printing notes.
Reserve Requirements: Largely Abolished
The traditional check on money creation was the reserve requirement: banks had to hold a minimum fraction of deposits as reserves (either as vault cash or as deposits at the central bank). A 10% reserve requirement would theoretically limit the money multiplier to a factor of ten, every $1 of base money could support at most $10 in deposits.
In practice, reserve requirements have been steadily weakened. In March 2020, the Federal Reserve eliminated reserve requirements entirely for most US depository institutions. The UK, Australia, Canada, and many other developed economies had already moved to zero or near-zero requirements years earlier.
Banks today are constrained primarily by:
- Capital requirements: They must hold adequate equity relative to their risk-weighted assets (under the international Basel III framework), limiting how much they can lend relative to their own net worth
- Liquidity coverage ratios: They must hold enough liquid assets to survive a 30-day stress scenario without central bank support
- Profitability constraints: Banks only create money when they can find creditworthy borrowers willing to pay a profitable interest rate
- Central bank access: The Federal Reserve, the European Central Bank, and other central banks stand ready to lend to banks at the discount window, providing the ultimate backstop
The practical effect is that modern banks can create money as long as they can attract borrowers, maintain adequate capital, and remain confident in their access to central bank liquidity. There is no hard ceiling equivalent to a gold reserve.
Bank Runs: When the System's Vulnerability Is Exposed
Fractional reserve banking contains an inherent structural weakness: a bank's liabilities (deposits) are immediately callable, you can withdraw at any moment, while its assets (loans) are long-term and illiquid. Mortgages and business loans cannot be instantly recalled to pay depositors.
This mismatch is stable under ordinary conditions. It becomes catastrophic when confidence breaks down.
A bank run begins when enough depositors simultaneously decide to withdraw. Since no fractional reserve bank holds 100% of deposits, it cannot pay all depositors at once. As the word spreads that the bank may be in trouble, more depositors rush to withdraw before the vault empties, a self-fulfilling prophecy. The bank fails not necessarily because its loans are bad, but because its structure cannot withstand simultaneous demands.
History provides stark examples:
- US Banking Panic of 1907: A cascade of bank failures triggered by speculative lending and panic withdrawals, ultimately prompting Congress to create the Federal Reserve in 1913
- The Great Depression (1930-1933): Over 9,000 US banks failed, wiping out uninsured deposits and contracting the US money supply by roughly one-third, deepening and prolonging the depression
- Northern Rock (2007): The first British bank run in 150 years; depositors queued outside branches across the UK to withdraw savings after the bank's wholesale funding markets froze
- Silicon Valley Bank (2023): A modern variant, depositors withdrew $42 billion in a single day via mobile banking and social media, making it the fastest bank run in history
Deposit insurance: the FDIC in the United States (currently covering up to $250,000 per account), largely prevents retail bank runs by guaranteeing depositors won't lose their insured savings. But it does not eliminate the underlying structural mismatch. It transfers the risk to taxpayers and the broader financial system.
The Central Bank as Lender of Last Resort
Central banks exist, in significant part, to backstop fractional reserve banking and prevent self-fulfilling collapses from destroying otherwise solvent institutions.
The economist Walter Bagehot articulated the governing principle in his 1873 work Lombard Street: in a panic, the central bank should "lend freely at a penalty rate against good collateral." This is now known as Bagehot's Rule.
The logic is straightforward: if depositors know that a solvent bank can always obtain emergency liquidity from the central bank, they have less reason to panic and more reason to leave their deposits in place. The run never starts.
This lender of last resort function is the keystone that makes fractional reserve banking systemically survivable. Banks can operate with thin reserves because they know they can borrow emergency funds if depositors demand more cash than they hold. The Federal Reserve's discount window, the ECB's liquidity operations, and similar facilities at every major central bank serve this purpose.
The implicit consequence: when a large bank's risk-taking goes badly wrong, the costs are at least partially borne by taxpayers rather than the bank's shareholders alone. The phrase "too big to fail" describes the extreme case, an institution whose collapse would trigger systemic contagion, making bailout the least-bad option from a policymaker's perspective. This creates moral hazard: the incentive to take excessive risk because the downside is partially socialized.
The Inflationary Consequences of Credit Expansion
Credit-driven money creation connects directly to inflation. When banks collectively expand credit rapidly, the money supply grows faster than the production of goods and services. More money chasing the same goods pushes prices up, not dramatically in any single month, but steadily and cumulatively.
Ludwig von Mises and Friedrich Hayek, the foremost figures of the Austrian school of economics, argued that credit expansion beyond real savings does more than cause general price inflation. It generates artificial booms: periods of economic expansion fueled by newly created money that aren't backed by actual deferred consumption. Projects get funded that would never pencil out at honest interest rates. Asset prices rise beyond fundamental value. Workers and capital flow into sectors where demand was artificially inflated.
When the expansion reaches its limits, when borrowers can't repay, when credit standards tighten, when confidence wobbles, the contraction is painful and proportional to the preceding excess. Austrians call this the business cycle, driven not by some mysterious property of markets but by the distortions introduced by artificial credit expansion.
This framework explains the credit boom of the early 2000s with reasonable precision. Easy money, fractional reserve expansion, and securitization of mortgages channeled resources into housing construction and consumption. When the cycle turned in 2007, the contraction was severe. Central banks responded with quantitative easing, purchasing trillions in bonds to pump reserves into the banking system, and near-zero interest rates sustained for over a decade.
Why Bitcoin Is Designed Differently
Satoshi Nakamoto published the Bitcoin whitepaper in October 2008, at the precise moment when the structural failures of fractional reserve banking were impossible to ignore. The genesis block, mined in January 2009, contained an embedded message: "The Times 03/Jan/2009 Chancellor on brink of second bailout for banks." The juxtaposition was deliberate.
Bitcoin's design addresses each of fractional reserve banking's core structural vulnerabilities:
| Fractional Reserve Banking | Bitcoin |
|---|---|
| Money supply expands through credit creation | Fixed supply: 21 million BTC, enforced in code |
| Requires trust in banks and central banks | Trustless verification by any full node |
| Subject to bank runs (liabilities > liquid assets) | No fractional reserves; you hold what you hold |
| Vulnerable to government seizure or debasement | Private keys; no counterparty risk |
| Inflationary by design | Disinflationary; new supply halves every four years |
| Lender of last resort backstop required | No backstop needed; cannot create new BTC to bail anyone out |
Bitcoin does not eliminate credit. People can lend bitcoin to each other; Bitcoin-native financial services exist. But it separates money creation from credit expansion. The base money supply cannot be inflated by banks making loans or by governments facing fiscal pressure. New bitcoins are created only through mining, at a rate encoded in the protocol and known decades in advance.
The Lightning Network, Bitcoin's payment channel layer, introduces a form of credit relationship (you lock bitcoin into a channel and transact off-chain), but it doesn't create new base-layer bitcoin. The on-chain reserves remain verifiable.
A System That Works, Until It Doesn't
It is worth being precise: fractional reserve banking functions adequately for most people in stable economies most of the time. It channels savings into productive investment, enables mortgages and business formation, and facilitates commercial activity at a scale that commodity money alone could not support.
The costs emerge at the margins and in crisis. The structural mismatch between liquid liabilities and illiquid assets creates systemic fragility. The continuous creation of new money through lending produces a long-term erosion of purchasing power. And newly created money flows through the economy unevenly, through what economists call the Cantillon effect: benefiting those closest to money creation (banks, large borrowers, asset owners) before price increases diffuse through to ordinary wages and savings.
None of this is an argument for dismantling the banking system overnight. It is an argument for understanding how money actually works, and for recognizing why a growing number of people, particularly those who have lived through monetary crises, are drawn to a system built on a fundamentally different foundation.
Fractional reserve banking is a system of trust, backstopped by central authority, that works until trust fails. Bitcoin is a system of mathematics, backstopped by no one, that works regardless.
This guide is for educational purposes only and does not constitute financial advice. Economic and historical examples are described for informational context.