Money Creation: How Banks Actually Make Money Out of Nothing

This is the most important file in the knowledgebase. If you do not understand what is on this page, every other inflation chart you look at is a black box.

The core claim is simple, and it is no longer fringe: when a commercial bank makes a loan, it creates the money on the spot, by typing a number into the borrower’s deposit account. It does not lend out savers’ deposits. It does not lend out reserves. It does not need either before the loan is made. The Bank of England said so explicitly in its own Quarterly Bulletin in 2014 ¹. Richard Werner proved it empirically the same year, with a real loan from a real German bank, while watching the bank’s internal accounting in real time ². The textbook version of banking, taught in nearly every undergraduate economics course on Earth, is wrong about the most basic question in monetary economics: where does money come from.

What Money Actually Is in 2026

Money in a modern economy comes in layers.

Base money (M0). Money the central bank itself creates. Two components: physical cash (notes and coins) and central bank reserves (deposits commercial banks hold at the central bank). Reserves are the chips banks use to settle with each other on the central bank’s books. Ordinary people and businesses cannot hold reserves. Only banks can.

Broad money (M1, M2, M3). The money the rest of us actually use. The number on your bank statement, current accounts, savings accounts, money market deposits. Almost all of it consists of commercial bank deposit liabilities, IOUs from a private bank promising to give you cash on demand.

The Bank of England, in its 2014 Quarterly Bulletin, gives the proportions:

“The majority of money in the modern economy is created by commercial banks making loans.” ¹

“In the modern economy, most money takes the form of bank deposits. But how those bank deposits are created is often misunderstood: the principal way is through commercial banks making loans. Whenever a bank makes a loan, it simultaneously creates a matching deposit in the borrower’s bank account, thereby creating new money.” ¹

In the United Kingdom at the time of publication, roughly 97 percent of the broad money supply consisted of commercial bank deposits, with only about 3 percent in physical cash and central bank money available to the non bank public ¹. The proportions in the United States, the eurozone, Canada, Japan, and every other major fiat economy are similar. The number on your bank statement is not money the central bank printed. It is money a commercial bank typed into existence when somebody, somewhere, took out a loan.

Three Theories of Banking

Richard Werner, in his 2014 paper “Can banks individually create money out of nothing? The theories and the empirical evidence” published in the International Review of Financial Analysis, identifies three competing theories held during the past century ². They make incompatible empirical predictions. Only one is correct.

Theory 1: Financial intermediation theory (the modern textbook story)

Banks are middlemen. Savers deposit money. Banks aggregate those deposits and lend them to borrowers, earning the spread. The total money supply is unaffected by banking activity, because for every loan asset there is a deposit liability that came from somewhere else.

This is the version in nearly every introductory macroeconomics textbook from roughly 1990 onward. It is also implicit in most general equilibrium models used by central banks until very recently. It is wrong ² ¹.

Theory 2: Fractional reserve theory (older textbooks, also wrong)

Banks collectively (not individually) create money through a multiplier process. A deposit of 100 enters the system. The bank keeps 10 as reserves and lends out 90. The 90 is redeposited at another bank, which keeps 9 and lends out 81, and so on. Total expansion is 100 divided by the reserve ratio. The central bank controls the money supply by controlling base money; the banking system mechanically multiplies it.

This is what most older economists and journalists still believe. It is wrong, both empirically (the multiplier is unstable and after 2008 became literally meaningless) and causally (causation runs the other way: banks lend first, reserves are accommodated afterwards) ² ¹. The Bank of England states this directly:

“Another common misconception is that the central bank determines the quantity of loans and deposits in the economy by controlling the quantity of central bank money, the so-called ‘money multiplier’ approach. In that view, central banks implement monetary policy by choosing a quantity of reserves. And, because there is assumed to be a constant ratio of broad money to base money, these reserves are then ‘multiplied up’ to a much greater change in bank loans and deposits. For the theory to hold, the amount of reserves must be a binding constraint on lending, and the central bank must directly determine the amount of reserves. While the money multiplier theory can be a useful way of introducing money and banking in economic textbooks, it is not an accurate description of how money is created in reality.” ¹

Theory 3: Credit creation theory (correct)

Each individual bank creates money out of nothing when it makes a loan. The bank does not need prior deposits or prior reserves. It needs willingness to lend, an apparently creditworthy borrower, regulatory capital, and the ability to acquire reserves later for settlement. The deposit appears the moment the loan is granted because the deposit is the loan: the bank’s accounting entry creates them simultaneously, on opposite sides of its balance sheet.

This was the dominant view in economics roughly from the late nineteenth century to the 1930s (Henry Macleod, Knut Wicksell, Joseph Schumpeter, Irving Fisher, John Maynard Keynes in the Treatise on Money), was driven out of the textbooks during the postwar period in favour of the intermediation and fractional reserve stories, and has been empirically reinstated since 2014 ². Werner’s 2016 followup paper, “A lost century in economics: Three theories of banking and the conclusive evidence”, traces this disappearance and reinstatement in detail.

The Bank of England 2014 paper, written by Bank staff and published in the Bank’s own Quarterly Bulletin, sides explicitly with credit creation theory and against the other two.

The Bank of England’s Own Words

The Bank of England Quarterly Bulletin 2014 Q1 paper, “Money creation in the modern economy”, by Michael McLeay, Amar Radia and Ryland Thomas, is the central bank of the United Kingdom, in its own publication, telling its own staff and the public, that the textbook is wrong ¹.

Key passages, verbatim:

“Money creation in practice differs from some popular misconceptions: banks do not act simply as intermediaries, lending out deposits that savers place with them, and nor do they ‘multiply up’ central bank money to create new loans and deposits.”

“Whenever a bank makes a loan, it simultaneously creates a matching deposit in the borrower’s bank account, thereby creating new money.”

“The reality of how money is created today differs from the description found in some economics textbooks: rather than banks receiving deposits when households save and then lending them out, bank lending creates deposits.”

“In normal times, the central bank does not fix the amount of money in circulation, nor is central bank money ‘multiplied up’ into more loans and deposits.”

These quotes are not paraphrased from a critic. They are the operative paragraphs of a paper written by Bank of England economists and published by the Bank itself. The content is corroborated by similar admissions from the Deutsche Bundesbank in its April 2017 Monthly Report (“the financing power of banks is not constrained by the existence of central bank balances”), the Bank for International Settlements in multiple working papers (notably Borio and Disyatat), and the ECB in less prominent publications.

Textbook vs Reality: Side by Side

Question	Textbook (intermediation / fractional reserve)	Reality (credit creation, BoE 2014, Werner 2014)
Where do bank loans come from?	From deposits placed by savers, or from reserves multiplied through the system.	They are typed into existence the moment the loan contract is signed. No prior funds are required. ¹ ²
What constrains lending?	The deposit base or the reserve ratio.	Regulatory capital (Basel III), perceived creditworthiness of the borrower, profitability, and the bank’s risk appetite. Reserves are not the binding constraint. ¹
Direction of causation?	Reserves cause loans (multiplier).	Loans cause deposits, deposits cause demand for reserves, central bank accommodates. ¹ ³
Is the money supply exogenous (set by the central bank)?	Yes.	No. It is endogenous, driven by bank lending decisions. The central bank influences the price (interest rate), not the quantity. ¹ ⁴
What does QE do?	Increases the money supply by some multiple of the asset purchases.	Creates reserves that sit on bank balance sheets. Reserves are not lent out. The effect on broad money is indirect, through asset prices and lending incentives. ¹
Does saving cause investment?	Yes, savings are pooled and lent to investors.	Causation reversed. Investment is funded by newly created bank credit; savings adjust afterwards. ²
Where does new money first appear?	In productive sectors via savers and entrepreneurs.	Wherever banks find it most profitable to lend, which since 1980 has been overwhelmingly into existing assets (mortgages, leveraged buyouts, financial speculation). ² ⁵

The two columns are not “different perspectives”. They make different empirical predictions and one of them is false.

Reserve Requirements Have Mostly Been Abolished

The “fractional” in “fractional reserve banking” was already misleading even when reserve ratios were positive, because the constraint never bound the way the textbook said. As of 2026, reserve requirements have been removed in most of the developed world. The textbook story is not just analytically wrong, it is mathematically impossible: you cannot multiply by a fraction when the fraction has been set to zero.

United States. Reserve requirements for all depository institutions were reduced to 0 percent effective 26 March 2020 by the Federal Reserve Board. The Fed’s own page states: “the Board reduced reserve requirement ratios to zero percent effective March 26, 2020. This action eliminated reserve requirements for all depository institutions.” The change was framed as a response to the COVID emergency. It has not been reversed as of 2026 ⁶.
United Kingdom. Statutory reserve requirements were abolished in 1981. From 1981 to 2009, individual commercial banks set their own monthly voluntary reserve targets in private contracts with the Bank of England ⁷.
Canada. Reserve requirements were phased out over a two year period beginning June 1992. Canadian chartered banks have operated with zero statutory reserve requirements since 1994 ⁸.
Australia, New Zealand, Sweden, Hong Kong. All operate without statutory reserve requirements.
Eurozone. Maintains a 1 percent minimum reserve requirement, but in a system flooded with excess reserves from QE, this is not a binding constraint on any individual bank’s lending.

The binding regulatory constraint on bank lending in 2026 is capital, not reserves. Basel III sets minimum ratios of regulatory capital (CET1, Tier 1, Total Capital) to risk weighted assets. A bank that wants to extend a new loan must have enough capital to support the additional risk weighted asset. But capital is not a constraint on money creation in the way reserves were once imagined to be: a bank can issue new equity, retain earnings, or write loans against existing capital headroom. The capital constraint is real but limited and lagging. Within those limits, the bank still creates the deposit out of nothing the moment it grants the loan.

The Chain of Money Creation, Step by Step

What actually happens when a bank lends, in order. Each step is documented in the BoE 2014 paper and in Werner 2014.

Step 1: The lending decision. A bank evaluates a loan application: borrower income, collateral, business plan, credit history, prevailing interest rate environment. It checks its own capital headroom and risk appetite. It says yes ¹.

Step 2: The accounting entry. The bank simultaneously credits the borrower’s deposit account by the amount of the loan, and books a corresponding asset (the loan itself) on its balance sheet. No funds are transferred from any other account. No reserves are checked beforehand. The deposit did not exist a moment ago. It exists now ².

A T account showing the bank’s balance sheet before and after a new 200,000 loan to a new customer:

	Assets		Liabilities
Cash and central bank reserves	100	Existing customer deposits	1,000
Existing loan portfolio	900	+ Borrower’s new deposit (the loan proceeds)	+ 200,000
+ New loan asset (claim on borrower)	+ 200,000	Equity	100
Total	201,000	Total	201,100

(Numbers illustrative; the slight asymmetry reflects existing equity and is not load bearing.)

The crucial observation: both sides of the balance sheet expanded simultaneously by 200,000. Total deposits in the banking system rose by 200,000. Total loans rose by 200,000. New broad money has been created. Nothing came from anywhere else. This is what “out of nothing” means ².

Step 3: The borrower spends the money. The borrower writes a cheque, sends a wire, makes a card payment. The deposit moves to a different bank.

Step 4: Interbank settlement. The receiving bank now has a claim on the lending bank. They settle through the central bank’s reserve accounts. If the lending bank has insufficient reserves, it borrows them: from another bank in the interbank market, from a money market fund via repo, or, as a last resort, from the central bank’s standing lending facility. The central bank does not refuse. Lender of last resort is its core function ¹ ³.

This is the meaning of “loans cause reserves, not the reverse”. The bank made the loan first. The reserves were acquired afterwards, at the prevailing policy rate, and the central bank accommodated the demand because to refuse would force a contraction of the entire banking system.

Step 5: The borrower repays. When the loan is eventually repaid, the process runs in reverse. The deposit is debited; the loan asset is written down. Money is destroyed. This is why aggregate broad money can shrink during deleveraging episodes (1931 to 1933 in the United States, 1990 to 2002 in Japan, 2008 to 2010 globally), even when the central bank is desperately trying to expand it.

Why This Matters for Inflation

If banks create money out of nothing when they lend, then the money supply is endogenous: it is driven by bank lending decisions, not by central planners at the central bank. Easy money policy lowers funding costs and encourages more lending; lending creates deposits; broad money expands. Tight money raises rates, suppresses lending, and broad money growth slows or reverses ¹ ⁴.

Two consequences for inflation:

The composition of new credit determines what kind of inflation you get. If new money is lent into productive enterprise (factories, R&D, infrastructure), output grows roughly in step with money, and consumer prices stay tame. If new money is lent against existing assets (residential mortgages, commercial real estate, leveraged buyouts), no new output is created, and the new money chases existing asset stock. Asset prices rise. House prices rise. Equity multiples rise. This is asset price inflation, and it does not show up in CPI ². Since approximately 1980, the share of bank lending going to non GDP, asset purchase transactions has risen dramatically in every major economy. (See 01-foundations/cantillon-effect.md.)
CPI inflation lags asset price inflation, sometimes by years or decades. New money enters at a specific point (the borrower’s bank account); it ripples outward at finite speed. Asset markets, being deep, liquid, and continuously priced, respond first. The real economy responds later, when the money eventually reaches firms and households that bid for goods and services. The 2009 to 2021 episode is the textbook case: massive monetary expansion, calm CPI, ferocious asset price inflation. Then in 2021 to 2024, with direct fiscal channels open (see below), CPI finally caught up.

QE Explained Simply

Quantitative easing is the process by which a central bank buys financial assets (typically government bonds and mortgage backed securities) from the private sector, paying with newly created central bank reserves. From the BoE 2014 paper:

“QE involves a shift in the focus of monetary policy to the quantity of money: the central bank purchases a quantity of assets, financed by the creation of broad money and a corresponding increase in the amount of central bank reserves.” ¹

Mechanics: the Fed (or BoE, ECB, BoJ) announces it will buy, say, $80 billion of Treasuries per month. It buys them from primary dealers, who are commercial banks. It pays by crediting the seller bank’s reserve account at the central bank. Reserves are created on the spot. The seller bank now holds reserves instead of Treasuries.

Crucial point: the new reserves do not directly enter the real economy. Households and firms do not get reserves. Only banks can hold them. The real economy effects of QE are indirect, working through three channels: (a) lower long term interest rates, encouraging borrowing and spending; (b) higher asset prices, producing a wealth effect; and (c) portfolio rebalancing, where banks and asset managers, deprived of safe yield, take on more risk.

This is why QE in 2008 to 2020 produced massive asset price inflation (every major equity market, every housing market, every bond market) but only modest CPI inflation. The new money sat at the top of the financial system. It pushed up the prices of things rich people own. It did not flow into wages or consumer goods at the same speed.

The Federal Reserve’s balance sheet (FRED series WALCL) tells the story numerically: roughly $0.9 trillion in mid 2008, expanded to about $4.5 trillion by 2014 through three rounds of QE, briefly tapered to about $3.8 trillion by 2019, then exploded in March 2020 to a peak of approximately $8.97 trillion in April 2022, and has been gradually reduced since under quantitative tightening, sitting in the mid $6 trillion range as of 2026 ⁹. Roughly a tenfold increase in seventeen years, in the world’s most important central bank balance sheet.

Direct Fiscal Monetisation, 2020 to 2022

The 2020 to 2022 episode was different from prior QE rounds, and the difference matters for understanding the inflation that followed.

In a standard QE round, the central bank buys bonds from banks. Banks get reserves. The new money mostly stays in the financial system. CPI is calm.

In 2020 to 2022, governments ran enormous fiscal deficits (the United States ran roughly $3.1 trillion in 2020 and $2.8 trillion in 2021), the Treasury issued bonds to finance them, the Federal Reserve bought essentially all of those bonds (Fed Treasury holdings rose from roughly $2.5 trillion in early 2020 to roughly $5.7 trillion by April 2022 ⁹), and the Treasury then deposited the proceeds into the bank accounts of households and businesses via stimulus checks (CARES Act, Consolidated Appropriations Act, American Rescue Plan), expanded unemployment insurance, the Paycheck Protection Program, and child tax credits.

This is the chain that bypasses the asset price only channel. Newly created money entered the real economy directly, in the deposit accounts of consumers, who spent it on goods and services. Headline CPI (BLS series CPIAUCSL) accelerated through 2021 and reached a 12 month rate of 9.1 percent in June 2022, the largest 12 month increase since November 1981 ¹⁰. Eurozone HICP peaked at 10.6 percent in October 2022. UK CPI peaked at 11.1 percent in October 2022.

Distinguishing these two channels (QE into the financial system versus fiscal monetisation into the real economy) is essential. Both are “money printing”, but their effects on consumer prices differ. The first inflates assets; the second inflates everything.

Werner’s Empirical Experiment

The three theories make different predictions about what happens inside a bank’s accounting system when a single, real loan is granted. For a century, no economist had run the obvious experiment: borrow money from a real bank while watching its books.

In August 2013, Richard Werner did exactly this ².

Setup. Werner approached Raiffeisenbank Wildenberg eG, a small cooperative bank in Lower Bavaria, Germany, and arranged to take out a personal loan of 200,000 euros. Crucially, he obtained the bank’s prior agreement that he would have access to the relevant internal accounting and that the bank’s director would document, in writing, exactly what was done at each step.

The loan. On 7 August 2013, the loan contract was signed and 200,000 euros was credited to Werner’s account at the bank.

The decisive observation. The bank director confirmed in writing that:

No funds were withdrawn or transferred from any other internal account at Raiffeisenbank Wildenberg.
No funds were borrowed from any other bank or institution.
The bank did not check, before or during the granting of the loan, whether it held sufficient reserves at its central bank or its central institution.
No related transactions, transfers, or account bookings of any kind took place to “finance” the credit balance.

The 200,000 euros appeared on Werner’s deposit account purely through the accounting entry: a loan asset booked on the bank’s left side, a matching deposit liability booked on the right side, and the borrower’s account credited as a result.

The conclusion. This is exactly what credit creation theory predicts and what the other two theories say cannot happen. The financial intermediation theory predicts the bank must source funds from prior savings; this did not happen. The fractional reserve theory predicts that an individual bank cannot create money on its own and must rely on a system wide multiplier process; the bank in question created 200,000 euros with no system wide process at all. Werner’s paper concludes:

“It has been shown empirically that banks individually create money out of nothing.” ²

This was, in effect, the first time in the history of monetary economics that the question was settled by direct observation rather than indirect inference. It was published in a peer reviewed journal (International Review of Financial Analysis, December 2014). It has not been refuted. It is corroborated by the Bank of England’s 2014 paper and by the Bundesbank’s 2017 statement. The textbook is wrong. The bank really does just type the money in.

What Follows From All Of This

Three implications carry forward:

Money supply is endogenous. Determined by profit seeking commercial banks, with the central bank setting the price. Charts of M2, broad money, and bank credit are charts of bank behaviour, not central bank decree.
Asset price inflation precedes consumer price inflation. New money enters at a specific point and ripples outward, and since 1980 banks have lent overwhelmingly into existing assets. Decades of “low CPI inflation” coexisted with extreme inflation in housing, equities, and bonds.
The post 1971 system has no anchor. With reserve requirements at zero in most of the developed world, capital constraints lagging and procyclical, and a central bank that accommodates demand for reserves, there is no hard ceiling on broad money creation. The only effective constraints are borrower demand for loans and bank willingness to lend. Both expand during booms and collapse during busts, which is why the post 1971 system has produced the largest credit cycles in recorded history.

Banks create money out of nothing when they make loans. The central bank accommodates. Reserve requirements have been abolished. Capital constraints lag. The textbook is wrong. The Bank of England’s own staff said so in 2014. Werner proved it the same year with a real loan and a stopwatch. Read those two papers if you read nothing else.

Sources

McLeay, M., Radia, A. and Thomas, R. (2014). “Money creation in the modern economy”. Bank of England Quarterly Bulletin, 2014 Q1, pp. 14 to 27. https://www.bankofengland.co.uk/quarterly-bulletin/2014/q1/money-creation-in-the-modern-economy ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹ ↩¹² ↩¹³ ↩¹⁴ ↩¹⁵ ↩¹⁶ ↩¹⁷
Werner, R. A. (2014). “Can banks individually create money out of nothing? The theories and the empirical evidence”. International Review of Financial Analysis, 36, pp. 1 to 19. https://www.sciencedirect.com/science/article/pii/S1057521914001434 . See also Werner, R. A. (2016), “A lost century in economics: Three theories of banking and the conclusive evidence”, International Review of Financial Analysis, 46, pp. 361 to 379. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹ ↩¹² ↩¹³
Goodhart, C. A. E. (1984). “Monetary Theory and Practice: The UK Experience”. Macmillan. Goodhart’s writings throughout the 1980s and 1990s established that central banks accommodate demand for reserves and do not control them exogenously, contrary to the money multiplier story. ↩ ↩²
Borio, C. and Disyatat, P. (2011). “Global imbalances and the financial crisis: Link or no link?”. BIS Working Paper No 346. Bank for International Settlements. Argues that bank credit, not savings imbalances, drives global liquidity. https://www.bis.org/publ/work346.htm ↩ ↩²
Richard Cantillon, “Essai sur la Nature du Commerce en Général” (written ca. 1730, published 1755). The original analysis of how the entry point of new money into an economy determines who benefits from monetary expansion. Treated in detail in 01-foundations/cantillon-effect.md of this knowledge base. ↩
Federal Reserve Board, “Reserve Requirements”. Official statement: “the Board reduced reserve requirement ratios to zero percent effective March 26, 2020. This action eliminated reserve requirements for all depository institutions.” https://www.federalreserve.gov/monetarypolicy/reservereq.htm ↩
Bank for International Settlements, “Currency areas: United Kingdom” page, and Bank of England Quarterly Bulletin records on the abolition of statutory reserve ratios in 1981. https://www.bis.org/mc/currency_areas/gb.htm ↩
Clinton, K. (1997). “Implementation of Monetary Policy in a Regime with Zero Reserve Requirements”. Bank of Canada Working Paper 97 to 8. Documents the 1992 to 1994 phase out of reserve requirements in Canada. https://www.bankofcanada.ca/wp-content/uploads/2010/05/wp97-8.pdf ↩
Federal Reserve Bank of St. Louis, FRED database, “Assets: Total Assets: Total Assets (Less Eliminations from Consolidation): Wednesday Level (WALCL)”. Pre crisis level approximately 0.9 trillion in mid 2008, peak of approximately 8.97 trillion in April 2022. https://fred.stlouisfed.org/series/WALCL ↩ ↩²
U.S. Bureau of Labor Statistics, “Consumer Price Index Summary”, news release for June 2022, published 13 July 2022. “The all items index increased 9.1 percent for the 12 months ending June, the largest 12-month increase since the period ending November 1981.” https://www.bls.gov/news.release/archives/cpi_07132022.htm . See also CPI home: https://www.bls.gov/cpi/ ↩