1Commit to a prediction
Imagine an economy divided into ten groups of people with equal incomes. New money equal to 10 percent of the existing money supply enters the economy through Group 1. Group 1 spends it, which raises the income of Group 2, and so on, until the new money has reached all ten groups. By the end, every income and every price has risen by 10 percent.
Over the whole period while the money is spreading, which outcome do you expect?
2Trace where new money enters
New money does not arrive in everyone's account at once. It enters through specific channels, and each channel has identifiable first recipients. Three channels account for most new money in modern economies.
Commercial bank lending
When a commercial bank makes a loan, it credits the borrower's account with a new deposit. That deposit is new money. The Bank of England has described this as the principal way money is created in modern economies.
First recipients are borrowers, and the sellers of whatever they buy, such as homes and businesses.
Central bank asset purchases
A central bank can buy government bonds and other financial assets and pay for them with newly created reserves. Large programs of this kind, called quantitative easing, were used widely after 2008 and again in 2020.
First recipients are banks and investment funds that sell the assets, and owners of financial assets whose prices rise.
Government borrowing
When a government spends more than it collects in taxes, it borrows by selling bonds. When banks or the central bank buy those bonds, new money is created and flows out through government spending.
First recipients are government contractors, employees, and people who receive government payments.
3Model the spread of new money
The model follows ten groups, each earning one unit of income per month. New money reaches Group 1 in the first month and then reaches one additional group at a fixed interval. When a group is reached, its monthly income rises by the size of the injection. The price level rises in step with the share of groups already reached, so that total spending always buys the same total quantity of goods. The model measures each group's real income, meaning what its income can actually buy at current prices.
Loading the Python engine. This takes several seconds on the first visit.
Each bar is the total change in a group's real income over the whole period, measured in months of its original income.
Real income of the first, middle, and last groups over time, as a percentage of original purchasing power. The dashed line is the price level.
4Reconcile the model with your prediction
Lock in a prediction in Step 1, then run the model to compare.
Why do early recipients gain?
Group 1 spends its new money while prices have barely begun to rise, so the money buys nearly its full original value. By the time the money reaches Group 10, prices have already risen, but Group 10's income has not. For many months Group 10 pays higher prices out of an unchanged income. The effect is named after Richard Cantillon, an eighteenth-century economist who first described it.
Why do the gains and losses cancel?
New money does not create new goods. In the model, the economy produces the same quantity of goods every month, so any extra goods one group buys must come from another group buying fewer. The injection therefore redistributes purchasing power without increasing the total. Once the money has reached every group, incomes and prices have risen by the same percentage and no group is better or worse off than before. The redistribution happens entirely during the transition.
How large is this effect in real economies?
Economists agree that the channel of entry shapes who benefits first, but they disagree about the size and importance of the effect. Research on quantitative easing, for example, finds that it raised the prices of stocks and bonds, which benefits people who already own those assets. Other research finds that it also lowered unemployment, which benefits working households. Whether the net result widened or narrowed inequality remains an open empirical question. This model isolates the timing effect and ignores those other channels by design.
What does this model leave out?
The model assumes a fixed quantity of goods, a single fixed path for the money, prices that respond immediately to spending, and no borrowing, interest, or changes in employment. Real economies violate each of these assumptions. Consider which assumption, if relaxed, would most change the result.
5Inspect the code
Nothing on this page is hidden. The Python below is the exact code that produced your results, running in your browser through Pyodide.
Show the Python model
Related reading. A short history of currency debasement traces this pattern from ancient Rome to the modern dollar.
Previous module, Purchasing power over time Next module, Reserve currencies and global trade