One dollar that spends itself many times
The earlier guides in this rung gave you the levers: government spending and taxes, the two halves of the budget. Now we ask the question that decides whether those levers are powerful or feeble — how much does one extra dollar of public spending actually change total output? You met the seed of the answer back in the business-cycle rung as the [[multiplier-effect|multiplier effect]]. The [[fiscal-multiplier|fiscal multiplier]] is simply that idea applied to the government's budget: it is the ratio of the change in GDP to the change in fiscal policy that caused it. A multiplier of 1.5 means a dollar of extra spending raised output by a dollar and a half.
How can a dollar become more than a dollar? Follow it. The state pays a builder a dollar to fix a road. That dollar is now the builder's income — and the builder does not bury it. Suppose she spends 80 cents of it (her marginal propensity to consume) on groceries. Those 80 cents become the grocer's income; he spends 80% of that, 64 cents, at the barber; the barber spends 80% of *that*, and so on. Each round is smaller because some money leaks out into saving and taxes, but the rounds keep going. The original dollar circulates through many hands, and the total spending it sets off adds up to far more than the single dollar the government put in.
Round 1: $1.00 (government pays the builder) Round 2: $0.80 (builder spends 80%) Round 3: $0.64 (grocer spends 80% of 0.80) Round 4: $0.51 (barber spends 80% of 0.64) ... ... (each round = 0.8 x the last) -------------------------------------------- Total = 1 / (1 - 0.8) = 1 / 0.2 = $5.00 Multiplier = 1 / (1 - MPC), here MPC = 0.8 -> 5
Stabilizers that work while no one is watching
Before we let the skeptics in, meet a quieter cousin of deliberate stimulus. Much of fiscal policy does not wait for a politician to act at all. [[automatic-stabilizers|Automatic stabilizers]] are features already baked into the budget that lean against the wind on their own. When a recession hits, incomes fall, so people pay less in income tax — and because most systems are progressive, their tax bill falls by *more* than their income does. At the same time, more people claim unemployment benefits and other support. Spending rises and taxes fall automatically, cushioning demand, with no new law required.
Their genius is timing. Deliberate, discretionary stimulus suffers from lags — the slump must be recognized, a bill debated and passed, the money actually spent — and by the time the road gets built the recession may be over, so the boost lands in the wrong year. A stabilizer fires the instant incomes turn, with no recognition lag and no legislative lag. In a boom it works in reverse: rising incomes push people into higher tax brackets and fewer people claim benefits, gently cooling an overheating economy. It is fiscal policy on autopilot, smoothing the business cycle in both directions.
The skeptic's reply: crowding out
Now the counter-story, and it is a serious one. The government usually pays for extra spending not from a piggy bank but by borrowing — it sells bonds, adding to the national debt. But the pool of savings people are willing to lend is finite. When the state steps in to borrow a large slice of that pool, it competes with firms that also want to borrow to build factories and buy machines. More demand for the same loanable funds pushes the interest rate up, and a higher interest rate makes some private investment projects no longer worth doing. The public dollar borrowed displaces a private dollar that would otherwise have been invested. Economists call this [[crowding-out|crowding out]].
How much gets crowded out is the whole ballgame. If a dollar of public borrowing displaces a full dollar of private investment, the multiplier collapses toward zero — the government merely rearranged who does the spending, adding nothing. If it displaces nothing, the multiplier stays large. The honest answer is that it depends sharply on the state of the economy. In a deep slump with idle factories, unemployed workers, and an interest rate already pinned near zero, there is little private investment to crowd out and plenty of slack to fill, so the multiplier tends to be large. Near full employment, with resources already in use, the same spending mostly bids resources away from private hands and crowds out more — the multiplier shrinks.
There is a subtler displacement too. Under the idea of [[ricardian-equivalence|Ricardian equivalence]], forward-looking households see today's borrowing as tomorrow's tax bill and save more now to pay for it, so the extra public spending is offset by reduced private spending — crowding out through expectations rather than interest rates. In its pure form this requires almost superhuman foresight and is rarely seen in full, but it points at a real truth: how people *expect* a deficit to be repaid shapes how much it actually stimulates.
Why the size is genuinely disputed
So what is the multiplier, really? Here you must hear the field honestly: there is no single number, and the disagreement is not laziness — it is one of the hardest measurement problems in macroeconomics. Empirical estimates range from below 0.5 (mostly crowded out, even contractionary in some readings) to above 2 (strongly stimulative), and credible economists land in different places. The reason it is so hard is that we never see the counterfactual. To know the multiplier we would need to compare the actual economy against the same economy *without* the spending — and that parallel world does not exist, so every estimate leans on a model and on assumptions that cannot all be checked.
Worse, the multiplier is not even one fixed quantity waiting to be found — it genuinely *changes* with circumstances. The same dollar has a different effect depending on whether the economy is slack or tight, whether the central bank offsets the stimulus by raising rates or accommodates it, whether the money is spent at home or leaks abroad through imports, and on what exactly it is spent (a temporary tax rebate that people save behaves very differently from a permanent infrastructure program). A number that is honestly 0.6 in a boom can be honestly 1.8 in a depression. Anyone who quotes you one universal multiplier is selling certainty the evidence does not support.
Holding both stories at once
The mature view is not to pick a side but to see that the multiplier and crowding out are two forces acting on the same dollar at the same time. The net effect — what actually happens to GDP — is the multiplier's push minus crowding out's drag. When the economy is deep in a slump, the push is strong and the drag is weak, so deficit-financed spending can genuinely raise output. When the economy is running hot, the drag overwhelms the push, and the same policy may simply raise interest rates and crowd out private activity while adding to the debt. Same tool, opposite results, depending on when you reach for it.
This is also where fiscal and monetary policy meet. Whether crowding out bites depends heavily on the central bank: if it lets interest rates rise to choke off the stimulus, crowding out is severe; if it holds rates down — especially at the zero lower bound you met in the monetary rung — the drag nearly vanishes and the fiscal multiplier swells. So the budget lever you have studied in this rung never acts alone. It is always pulling against, or working with, the monetary policy lever from the rung before — and the two together, not either by itself, decide what one public dollar finally does.