Welcome to your study space
Understand economics at your own pace: step-by-step guided courses, and exercises to practice.
The causes of inflation
Demand, costs, money: where do price rises really come from?
M·V = P·YProblem / motivation
Measuring inflation tells you how much prices are rising — not WHY they are rising. And without a diagnosis, there is no suitable remedy.
Economists distinguish three broad families of causes. DEMAND-pull inflation: too much spending against a supply — everything firms are able to produce — that cannot keep up; prices act as the safety valve, this is an economy “overheating”. COST-push inflation: energy, raw materials or wages become dearer, and firms pass it on to their selling prices. MONETARY inflation, finally: when the quantity of money in circulation grows durably faster than output, prices end up following.
History supplies a textbook case per family, in the same order. Demand: the United States in 2021-2022, pumped up by stimulus packages — the government hands out massive support, spending leaps. Costs: the oil shocks of the 1970s — oil suddenly four times dearer — which pushed the rise in French prices beyond 13% a year in late 1974. Money: Germany in 1923, where prices doubled in a few days. The 2021-2023 surge in the euro area — the countries sharing the euro — mixed demand and costs, energy first; the share taken by money remains debated. These three families are the model of this course; like any model, it rests on a few assumptions — see if you can guess them first.
Assumptions
In your view, which assumptions are needed for this model to hold? Jot down your ideas — no lead is wrong, this is your worksheet.
Your worksheet is still empty. Go for it: propose at least one idea.
0 idea(s) proposedFormalization
Demand and costs need no equation: a causal chain is enough. Demand — spending leaps, order books fill up, delivery times lengthen, and sellers, sure of selling, raise their prices (that is assumption 3 in action). Costs — energy or wages become dearer, margins (what the firm earns once its costs are paid) get squeezed, it passes this on to its prices to survive. (The course “Shocks and equilibrium AS-AD” — aggregate supply, aggregate demand — puts these two chains into a graph.)
The monetary cause, on the other hand, can be put into an equation. Follow the euros of an imaginary country for a year: if it holds M = €1,000 billion of money (notes, current accounts) and each euro is used on average V = 4 times in the year, total spending for the year is M × V = €4,000 billion.
That same spending buys everything that is sold — and in our toy country, everything bought was produced during the year: Y, the quantities produced, that is 500 billion “baskets” (a fictional unit of output — nothing to do with the CPI basket), at an average price P of €8 each. P × Y = €4,000 billion as well. A single flow of purchases, counted twice: there is the “equation of exchange” M × V = P × Y. It reads like a sentence — money multiplied by its rate of use equals prices multiplied by quantities — and it is always true, by construction: an accounting identity, not yet a theory.
For it to EXPLAIN inflation, our assumptions are needed: if velocity V is stable (assumption 5) and if, in the long run, output Y depends on factories and hands — not on the quantity of money (assumption 2, long-run side) — then when M swells durably, the only term that can follow is P. This is the quantity theory of money; the course of that name (Money & banking module) argues it out step by step. Click each term:
Solving / calculation
From one year to the next, each letter of the equation moves by a small percentage — written g( ), g for growth: g(M) = 6 means M gains 6% in the year. The bridge from the course “Inflation” (adding x% means multiplying by 1 + x/100) then gives the rule of the game: when quantities MULTIPLY, their growth rates ADD UP — give or take a whisker: ×1.06 on M and ×1.02 on V make ×1.0812 on the product M·V, very nearly +8%. (That is the whole point of the ≈ that follows: an approximation, excellent for moderate rates, increasingly wrong when they run away — for Germany in 1923 it is worth nothing at all.) Let's unroll it, then take the controls.
- The two sides of M·V = P·Y, equal this year, stay equal next year: they grow at the same pace — and on each side, the rates add up
g(M) + g(V) ≈ g(P) + g(Y) - g(P), the speed at which prices rise, is exactly π — the inflation rate of the previous course
g(M) + g(V) ≈ π + g(Y) - We isolate inflation
π ≈ g(M) + g(V) − g(Y) - Money grows by 6%, velocity stable (assumption 5), output +2%
π ≈ 6 + 0 − 2π ≈ 4%
Reading the signs: output ABSORBS the monetary push (the − in front of g(Y)); velocity, for its part, works BOTH ways — it adds on when money circulates faster, it absorbs when money sleeps (g(V) negative). At stable velocity, the excess growth of money over that of output ends up in prices. This is a LONG-run reading: the equation is always true, but the CAUSALITY money → prices takes years and rests on our assumptions — it says nothing about next quarter.
π ≈ g(M) + g(V) − g(Y)The simulator takes the toy country of the formalisation (M = €1,000 billion, V = 4, Y = 500 billion baskets, average price €8). Set the three growth rates: it recomputes the price EXACTLY through P = M·V ÷ Y, then compares with the shortcut.
Economic interpretation
Three families, three remedies — the whole art of the central banker (the pilot of the central bank, the institution that issues the currency and watches over prices: the ECB in the euro area) is diagnosis. And one criterion settles everything: what makes a rise DURABLE?
When spending exceeds what the economy can produce (assumption 3), braking calms prices: the central bank raises its policy rate — the rent on money, which makes credit dearer and cools spending (course “The policy rate and transmission”) — and the government moderates its budget. At the cost of a slowdown. Textbook case (debated in detail, as always): the United States in 2021-2022, pumped up by stimulus packages.
Faced with oil doubling, raising rates does not bring the barrel down: the central bank can only stop the shock from SETTLING IN. Settling in how? Through the price-wage loop: prices rise → employees obtain increases to keep up → firms' costs rise → they raise their prices → and so on. France in the 1970s, with wages then indexed to prices — revalued automatically at every rise — lived through it on a grand scale: more than 13% inflation a year in late 1974, until de-indexation in 1983 cut the loop (only the minimum wage remains indexed today). Another cost coming from elsewhere: the exchange rate — energy is paid for in dollars, and a weak euro makes the bill heavier (course “What is an exchange rate?”).
The American economist Milton Friedman popularised the idea that DURABLE inflation is first of all a monetary phenomenon — Germany in 1923 remains its extreme illustration: the state paid its spending with money created for the occasion (the “printing press”), and prices doubled in a few days. The course “The quantity theory of money” (Money & banking module) argues this reading and its conditions out step by step.
An isolated shock makes a jolt: prices jump, then the rise dies out once the shock has been digested. Inflation becomes DURABLE only if some engine sustains it — money going along with it, or the price-wage loop fed by expectations (assumption 4). Keep this criterion for the exercises; “The Phillips curve”, the last course of the module, turns it into a theory.
Limits / critiques
Exercises
The money supply grows by 8% a year, velocity is stable, output grows by 3%. What long-run inflation should you expect (in %)?
True or false: on the monetary reading, money creation that is durably faster than output mainly increases output in the long run.
The money supply grows by 10% a year, output by 1%, but the velocity of circulation FALLS by 6% (crisis hoarding: money sleeps). What long-run inflation does the equation of exchange predict (in %)?
The ECB aims for 2% inflation. Output grows by 1.5% a year and velocity is stable. What growth of the money supply is compatible with the target (in %)?
The toy country has grown: M = €1,200 billion, V = 4, output Y = 600 billion baskets. What is the average price of the basket P (in €)?
Oil doubles and hauliers pass the rise on to their rates. Which family of inflation is this?
An economy at full employment — factories and hands already all busy: households draw on their savings, the government hands out support, spending leaps and order books fill up — prices rise. Which family of inflation?
True or false: a one-off oil shock is enough, on its own, to create durable inflation.
2021-2023 in the euro area: energy soars, order books fill up at the post-Covid reopening, governments support incomes. What is the diagnosis?
Hyperinflation: in one year the money supply is multiplied by 2 (+100%) and, with everyone getting rid of their notes as fast as possible, velocity gains 20%. Output stable. The shortcut predicts 100 + 20 − 0 = 120%. What is the REAL rise in prices (in %)?