Article · Money
Money printing: what happens when countries print money
Over long stretches, prices follow money. Across 151 countries, average money growth and average inflation line up with a correlation of 0.96: where money grew fastest, prices rose fastest. In the United States, money grew +26.8% in the year to February 2021, and consumer inflation peaked at +9.0% 16 months later. Yet at low rates the link is loose: between August 2008 and March 2021 the Fed's balance sheet grew ×8.4 while consumer prices rose 1.5% a year.
M2 grew $3.77T in 2020 (+24.5%); only 7% of it was paper currency
consumer inflation peaked 16 months later, at +9.0% (June 2022)
correlation of average money growth and average inflation in 151 countries, 1961–2025
April 13, 2022: ×9.9 its size before Lehman ($910B); $6.75T on November 4, 2026
total assets as a share of GDP in 2021, from 15% in 2007
Money and prices across countries
The quantity theory of money says that when money grows much faster than the goods it can buy, prices make up the difference. A simple test: take every country with at least 20 years of both numbers in the World Bank's data — 151 countries, 1961 to 2025 — and average each one's yearly growth of broad money (cash plus bank deposits) and its yearly consumer inflation over the same years.
The dots fall along a line. The correlation is 0.96, and on the chart's log scale the fitted line has a slope of 0.98: each extra point of money growth came with about 0.98 points of extra inflation. Fastest of all: Brazil (money 99.0%, prices 83.1%), DR Congo (money 91.5%, prices 86.2%) and Angola (money 64.8%, prices 61.5%). The United States: money 7.3% a year, prices 3.8% (1961–2024). Most dots (149 of 151) sit below the dashed line, where prices rose more slowly than money — by a median 6.4 points a year — because growing economies need more money for more goods, and people tend to hold more money as they get richer.
Near the low end the link is loose. Among the 28 countries whose money grew less than 10% a year on average, the correlation is 0.54 (against 0.96 for the other 123), and their inflation ranged from 0.5% to 6.7%. At low rates, other forces — output growth, interest rates, how much cash people choose to hold — move prices as much as money does. The slowest money growth: Brunei (money 4.2%, prices 0.5%) and Antigua and Barbuda (money 4.9%, prices 2.2%).
The same test, decade by decade
| Decade | Countries | Correlation |
|---|---|---|
| 1960s | 45 | 0.75 |
| 1970s | 76 | 0.91 |
| 1980s | 92 | 0.96 |
| 1990s | 114 | 0.95 |
| 2000s | 146 | 0.83 |
| 2010s | 148 | 0.78 |
| 2020s* | 114 | 0.83 |
Correlation of average money growth and average inflation (log scale) across the countries with at least 8 years of both in the decade. It stays between 0.75 and 0.96 in every decade. * The decade under way: countries with at least 5 years, to 2025.
The balloons: central banks after 2008 and 2020
Central banks create money by buying assets — mostly government bonds — and paying with new reserves, the deposits that commercial banks keep at the central bank. On a large scale this is called quantitative easing (QE), and a central bank's total assets are a good gauge of how much it has created. In 2007 the Federal Reserve held assets worth 6% of US GDP; after the rescues of 2008, three rounds of bond buying and then the pandemic, it reached 37% in 2021. The Eurosystem — the European Central Bank and the euro area's national central banks — went from 15% to 68% (2021).
Each balloon is one year's balance sheet as a share of that year's GDP: the higher and bigger, the more money the central bank had created. From 2007 to each one's peak: Federal Reserve ×6.0, Eurosystem (ECB) ×4.6, Bank of Canada ×7.2 and Reserve Bank of Australia ×3.5. All 4 have shrunk since their peaks, as the central banks let bonds mature without buying new ones (quantitative tightening). On November 4, 2026 the Fed's assets stood at $6.75T.
2007: 6% · peak 37% (2021) · 2025: 21%
2007: 15% · peak 68% (2021) · 2025: 39%
2007: 3% · peak 25% (2020) · 2025: 7%
2007: 9% · peak 30% (2021) · 2025: 14%
The balloons as a table
| Central bank | 2007 | Peak | Latest |
|---|---|---|---|
| 6% | 37%2021 | 21%2025 | |
| Eurosystem (ECB) | 15% | 68%2021 | 39%2025 |
| 3% | 25%2020 | 7%2025 | |
| 9% | 30%2021 | 14%2025 |
Where the money showed up: assets first
From August 2008 to March 2021 the Fed's balance sheet grew ×8.4. Consumer prices rose ×1.2 over those 13 years — 1.5% a year, close to the Fed's 2% goal. Asset prices moved much more: the S&P 500 ×3.1 and US home prices ×1.5 (both before inflation, the S&P without dividends).
Why did the money not reach the shops? Most of it stayed in the banking system as reserves, and much of the rest went into savings and markets rather than spending — the story of the case of the missing inflation. Consumer prices only woke up in April 2021, the first month after the pandemic began with inflation of 3% or more over twelve months.
2020: the printing press at full speed
In 2020 the US money stock M2 — cash plus checking, savings and money-market deposits — grew by $3.77T, +24.5% in a single year. Spread over the year's 366 days, that is $10.3B a day. The counter below runs at that pace from the moment you opened this page.
New US money since you opened this page, at 2020's pace
$0
M2 grew $119K a second in 2020
M2 added in each month of 2020
How much of it was actually printed
Paper currency and coins: $261B (7%)Bank deposits: 93%
M2 and its currency component, December 2019 to December 2020, seasonally adjusted monthly averages. Source: Board of Governors of the Federal Reserve System (H.6: M2SL, CURRSL), via FRED; population: BEA
The press was not steady: April 2020 alone added $1.03T, more than M2 grew in any whole year from 1960 to 2019. And “printing money” is mostly a figure of speech: only 7% of 2020's new money was paper. The rest was keystrokes — deposits created when the Fed bought bonds, when the Treasury sent relief payments and when banks lent. For comparison, from September 2025 to September 2026 M2 grew by $3.46B a day.
Money first, prices later
M2 grew +26.8% in the year to February 2021, the fastest since the series began in 1959. At that point consumer prices were rising only +1.7% a year. Inflation came later: it peaked at +9.0% in June 2022, 16 months after the peak of money growth. Then the story ran backwards. M2 reached $21.8T in March 2022, +41% above February 2020, and then shrank to $20.7T by October 2023 (−4.9%). It stayed below its level of a year earlier for 15 months in a row, from December 2022 to February 2024 — the first such falls since the series began in 1959. By September 2026 it was back up to $23.4T, +51% above February 2020.
Slide the money line forward in time to see the lag. Since January 2015, money growth matches inflation best when it is shifted 18 months: the correlation rises from −0.17 with no shift to 0.71. Over the whole record since January 1960 the best shift is 24 months, and the match is weaker (0.42).
Correlation 0.71
A correlation is not proof. Even at the best shift, money growth accounts for 50% of the swings in inflation since 2015 in the statistical sense (the square of the correlation), and other things moved at the same time: supply chains jammed, energy prices jumped in early 2022, and spending swung from services to goods. Before 2020 the best shift was 33 months and the correlation only 0.52. Money is a prime suspect, not a proven culprit.
The biggest printing sprees in the data
When a government pays its bills with new money for long enough, prices can run ahead of the money itself, because people rush to spend a currency they no longer trust. These are the 8 countries with the fastest one-year growth of broad money in the World Bank's data among the years that also have a consumer price figure, each in its record year; the bars show how many times money and prices multiplied. The full stories — and the cases these money figures miss — are in the hyperinflation article.
Bolivia 1985money ×71 (6,988%), prices ×118 (11,750%): prices outran money
DR Congo 1994money ×71 (6,969%), prices ×239 (23,773%): prices outran money
Peru 1990money ×65 (6,385%), prices ×76 (7,482%): prices outran money
Angola 1996money ×42 (4,106%), prices ×42 (4,145%): prices kept pace
Brazil 1993money ×34 (3,281%), prices ×20 (1,927%): money outran prices
Ukraine 1993money ×19 (1,809%), prices ×48 (4,735%): prices outran money
Azerbaijan 1994money ×12 (1,117%), prices ×18 (1,662%): prices outran money
Armenia 1994money ×8.4 (740%), prices ×35 (3,374%): prices outran money
The most famous cases fall outside these figures: Germany in 1923 and Hungary in 1946 (the highest inflation ever recorded) came before the data begin, and the money figures miss Zimbabwe in 2008 and Venezuela in 2018–19. Each followed years of budgets paid for with new money.
Questions people ask
Does printing money cause inflation?
Over long periods, mostly yes: across 151 countries, average money growth and average inflation have a correlation of 0.96, and the five countries with the fastest money growth all had inflation of 42.2% a year or more. Over short periods and at low rates the link is loose: among countries with money growth under 10% a year the correlation is 0.54. In the US, money growth of +26.8% in the year to February 2021 was followed by inflation peaking 16 months later, with other causes at work too.
What happens when a country prints too much money?
Prices rise, and if the printing goes on, people stop trusting the currency and spend it as fast as they can, which pushes prices up faster still. Among the years in the World Bank's data that have both a money and a price figure, the biggest one-year jump in broad money was in Bolivia in 1985: money ×71, consumer prices ×118. Such episodes usually end with a new currency or a peg to a foreign one, and a government that stops paying its bills with new money.
How much money did the Fed print?
The Fed's balance sheet — the money it created by buying bonds and lending — grew from $910B before Lehman Brothers failed (August 27, 2008) to $8.97T at its peak (April 13, 2022), ×9.9. Of that, $4.81T came after the pandemic began (from $4.16T on February 26, 2020). The wider money stock M2, which includes bank deposits, grew by $3.77T in 2020 alone; only $261B of that was paper currency.
What is QE?
Quantitative easing: a central bank buys large amounts of government bonds (in the US also mortgage-backed securities) and pays with newly created reserves, to push long-term interest rates down when the short-term rate is already near zero. The Fed ran QE in three rounds from late 2008 to 2014 and again from March 2020 to early 2022; at the peak its assets were 37% of US GDP (2021, December). Quantitative tightening (QT) is the reverse: letting bonds mature without replacing them.
Why didn't QE cause inflation at first?
Because most of the money did not reach the shops. QE created bank reserves that stayed in the banking system, banks lent cautiously after 2008, and people and firms held on to cash while interest rates were near zero. From August 2008 to March 2021 the Fed's assets grew ×8.4 while consumer prices rose 1.5% a year; asset prices rose much more. In 2020–21 new money went straight to households through relief payments while supply was short, and inflation followed. More in Unsolved mysteries of economics.
Go deeper
How the numbers are calculated
- Money against prices: for every country, the years with both broad money growth and consumer inflation in the World Bank's WDI (7,308 country-years). A country's dot needs at least 20 such years (151 countries); a country-decade at least 8 (5 in the decade under way). Averages are compounded: exp(mean of ln(1 + rate)) − 1. Correlations and the fitted line use ln(1 + rate), the chart's scale. Broad money is measured end of year to end of year, inflation as the change of the yearly average price level.
- Breaks left out: a year in which broad money fell by more than 50%, or multiplied more than 4 times as much as prices, signals a redenomination or a change of definition rather than money growth, and is left out (12): DR Congo 1983 (1,113%); DR Congo 1984 (−81.7%); Ecuador 1999 (−50.8%); Guinea-Bissau 1990 (575%); Guinea-Bissau 1991 (−67.3%); Guinea-Bissau 2003 (−57.6%); Equatorial Guinea 1990 (−52.0%); Grenada 2024 (−91.3%); Grenada 2025 (1,169%); Israel 1981 (829%); North Macedonia 1994 (−55.2%); Sierra Leone 2015 (−99.9%).
- Balloons: total assets in December — the average of the weekly figures for the Fed and the Eurosystem, the RBA's monthly average, the end-of-month figure for the Bank of Canada — divided by GDP of the same calendar year: US GDP from BEA (the average of the four quarterly annual rates), euro area GDP from the ECB (the sum of four quarters; 20 countries at a fixed composition, so in the early years it covers a few countries that had not yet joined the euro), Canada and Australia from the World Bank as published (national accounts there may follow a fiscal year). The Bank of Japan and the Swiss National Bank are not shown: their terms of use do not clearly allow republishing. Total assets of all Federal Reserve Banks, Wednesday level, not seasonally adjusted; weekly since 18 December 2002. Total assets of the Eurosystem (the ECB and the national central banks of the euro area, changing composition). The ECB dates each week by its ISO week; the date here is that week's Friday. Total assets at the end of each month. Months the Bank publishes as 0 before it opened in March 1935 are left out. Monthly averages of the weekly figures: to June 2013 as the RBA's historical table publishes them, from July 2013 computed by BigCapital from the weekly table.
- Assets after 2008: indexes with August 2008 = 100. The quiet years end the month before US consumer prices first rose 3% or more over twelve months after the pandemic began (April 2021).
- Lags: twelve-month changes of monthly M2 and CPI-U (both seasonally adjusted). The correlation at a shift of k months uses every month of the period in which inflation and money growth k months earlier both exist. Seasonally adjusted. The Fed changed the definition in May 2020 (savings deposits moved into M1 and stayed in M2). All items in US city average, all urban consumers, seasonally adjusted (BLS CUSR0000SA0). Months without a published value are left out (2025-10).
- Counter: M2 in December 2020 minus December 2019, divided by the 366 days of 2020; the counter multiplies the pace per second by the seconds since the page opened. Per American: divided by the BEA's mid-year population.
- Source: World Bank, World Development Indicators: Broad money growth (annual %), Inflation, consumer prices (annual %) and GDP (current local currency), CC BY 4.0 (original source of the money and price figures: IMF, International Financial Statistics); averages and ratios calculated by BigCapital.
- Source: Board of Governors of the Federal Reserve System: total assets (WALCL, H.4.1), M2 (M2SL, H.6) and its currency component (CURRSL), retrieved from FRED, Federal Reserve Bank of St. Louis; US GDP and population: U.S. Bureau of Economic Analysis, retrieved from FRED. Public domain.
- Source: European Central Bank, Eurosystem consolidated weekly financial statement, total assets and euro area GDP at current prices; December averages and ratios to GDP calculated by BigCapital.
- Source: Bank of Canada, Bank of Canada assets and liabilities: month-end, total assets (V36651); ratios to GDP calculated by BigCapital.
- Source: RBA, Statistical Table A1, Reserve Bank of Australia – Balance Sheet; monthly averages from July 2013 and ratios to GDP calculated by BigCapital.
- Source: U.S. Bureau of Labor Statistics, CPI-U, all items, seasonally adjusted (CPIAUCSL), retrieved from FRED, Federal Reserve Bank of St. Louis.
- Data: Robert J. Shiller, Online Data (Yale University): the S&P Composite index, monthly average price.
- Homes: Robert J. Shiller, US Home Prices 1890–Present, a chain of sources; since 1975 the S&P CoreLogic Case-Shiller U.S. National Home Price Index.
- Flags: flag-icons by Panayiotis Lipiridis (MIT License).
- This article describes data; it is not investment advice.
Reuse this article
The text and charts are free to republish, translate and adapt under CC BY 4.0: credit BigCapital.org with a link to this page. Data from other providers keeps its own license (see the sources below). How to credit
Source: BigCapital.org — “Money printing: what happens when countries print money”, https://bigcapital.org/articles/money-printing/