Fiscal Policy

Expansionary and Contractionary Fiscal Policy

Fiscal policy is Congress using its power to tax and spend to shift aggregate demand. That is the whole concept in one sentence.

The CARES Act of March 2020 is probably the clearest real-world example you will ever see. Congress authorized $2.2 trillion in spending, $1,200 checks to about 160 million people, $25 billion in airline payroll grants, $175 billion for hospitals, and roughly $800 billion through two rounds of the Paycheck Protection Program for small businesses. The economy had shed 22 million jobs in March and April, and GDP was contracting at an annualized 31.4% in Q2. Private spending had cratered, so the federal government stepped in as the spender of last resort. That is expansionary fiscal policy in its rawest form. Increase government spending (G), cut taxes (T), or both, and AD shifts right. Real GDP rises and the price level goes up too. The 2009 American Recovery and Reinvestment Act followed the same playbook during the Great Recession, $831 billion spread across infrastructure, state fiscal relief, and middle-class tax cuts.

Contractionary fiscal policy is the mirror image. If the economy is running past potential output and you have an inflationary gap, the textbook answer is to cut spending or raise taxes so AD shifts left. Straightforward on paper. But it almost never happens in reality, because elected officials who vote for tax hikes or gut popular programs tend to get voted out. You should know the theory cold for the AP exam, but also know that genuine contractionary fiscal policy is vanishingly rare in American politics.

The Spending Multiplier

When the federal government awarded $42 million to rebuild the I-35W bridge in Minneapolis after its 2007 collapse, that money did not just sit in one contractor's bank account. Flatiron-Manson hired welders, crane operators, and engineers. Those workers deposited paychecks and spent portions at grocery stores, restaurants, car dealerships around the Twin Cities. The grocery stores used that revenue to pay their employees, who turned around and spent again. Each round a bit smaller than the last because some fraction of every dollar gets saved instead of spent. That cascading chain is the multiplier effect.

The formula is simple. Spending multiplier = 1 / (1 - MPC). The marginal propensity to consume (MPC) is just the fraction of each additional dollar that people spend rather than save. MPC of 0.8 means people spend 80 cents and save 20 cents out of every new dollar. Plug that in: 1 / (1 - 0.8) = 1 / 0.2 = 5. So a $10 billion infrastructure bill would shift AD rightward by $50 billion as the initial spending cascades through round after round. If MPC jumps to 0.9, the multiplier is 10, each dollar bounces through more hands before leaking into savings. Higher MPC, bigger multiplier, more powerful stimulus per government dollar.

The Tax Multiplier

This shows up on the AP Macro exam constantly. Why does a $10 billion tax cut pack less punch than $10 billion in direct government spending?

Look at what happens in round one. Government spends $10 billion building highways, and all $10 billion enters the economy immediately because the government is literally buying something from someone. Tax cut of $10 billion? That lands in household bank accounts, but if the MPC is 0.8, households save $2 billion right off the top. Only $8 billion actually enters the first spending round. The chain starts smaller and stays smaller.

The formula is Tax multiplier = -MPC / (1 - MPC). Negative sign because taxes and AD move in opposite directions, a tax cut boosts AD, a tax hike shrinks it. With MPC = 0.8: -0.8 / 0.2 = -4. That $10 billion tax cut shifts AD right by $40 billion. Compare that to $50 billion from equivalent direct spending. The gap between the spending multiplier and the tax multiplier in absolute value is always exactly 1, regardless of the MPC. Always. It comes from that first-round leakage. Government spending injects 100% immediately while the tax cut loses the savings fraction before anybody spends a dime.

Crowding Out

The multiplier math assumes the government's spending just adds to total demand without any side effects. But deficit-financed stimulus has a complication.

When Congress spends more than it collects, the Treasury sells bonds to cover the gap. Those bonds compete for the same pool of loanable funds that businesses want for factory expansions, that developers need for housing projects, that families rely on for auto loans. More borrowers chasing a fixed pool of savings pushes interest rates up across the economy. A manufacturing firm that would have built a new plant at 4% shelves the project at 7%. A developer cancels a condo tower. A family decides the monthly car payment no longer works. Private investment drops, partially offsetting the government spending boost. That is the crowding-out effect, and it means the actual AD shift is smaller than the textbook multiplier would predict.

How severe is crowding out? It depends entirely on conditions. During the 2008 crisis, private investment demand had already collapsed and interest rates sat near zero. There was almost nothing to crowd out, so government spending faced minimal displacement. Contrast that with an economy humming along at full employment where firms are actively competing for every dollar of savings. In that environment, government borrowing can push rates up substantially and squeeze private investment hard. The theoretical extreme is full crowding out, where every government dollar displaces exactly one private dollar and the net AD shift is zero. But in practice, partial crowding out is the norm.

Automatic Stabilizers

Congress is slow. Committee hearings, floor debates, conference negotiations between House and Senate, presidential signature. The 2009 Recovery Act was considered fast and it still took nearly two months from inauguration day to signing. By the time money actually flows into the economy, the recession might have deepened or, awkwardly, already ended. Automatic stabilizers bypass all of that because they kick in without new legislation.

The progressive income tax is the big one. Worker loses her job or gets hours cut, taxable income drops, she falls into a lower bracket automatically. Disposable income does not crater as steeply as gross income did. During expansions, rising incomes push taxpayers into higher brackets, which pulls purchasing power out of the economy without any congressional vote. It acts as a natural brake on AD.

Unemployment insurance works the other side. Economy contracts, more workers qualify for benefits, no new law required. Claims surged from 211,000 per week in February 2020 to 6.9 million per week by late March, and those checks went overwhelmingly to people who would spend every dollar, cushioning consumption. SNAP enrollment follows similar logic, expanding automatically as incomes fall.

They dampen the swings. They do not end deep recessions by themselves. More like shock absorbers than engines.

Budget Deficits and the National Debt

In fiscal year 2020 the federal government spent roughly $6.55 trillion and collected about $3.42 trillion in revenue. That $3.13 trillion gap was a budget deficit, the largest in U.S. history in raw dollars, though the WWII deficits of 1943-1945 were proportionally bigger relative to GDP. Treasury covered the shortfall by selling bonds, notes, and bills to pension funds, foreign central banks, individual savers, commercial banks, and the Fed itself.

A deficit is a flow. How much red ink in one fiscal year. The national debt is a stock, the total pile of outstanding IOUs at any moment. Stack up every deficit ever run, subtract the occasional surplus (1998-2001 being the most recent stretch), and that is your national debt.

In theory the budget should roughly balance over a full business cycle. Deficits during recessions to prop up demand, surpluses during booms to rebuild fiscal capacity. In practice, surpluses are extraordinarily rare. Cutting spending and raising taxes is politically toxic no matter what the economy looks like. The U.S. ran deficits even during the long expansion of 2010-2019.

Whether big debt actually threatens the economy is one of the sharpest debates in the profession. Critics point to rising interest payments consuming more of the federal budget, the likelihood of future tax hikes to service obligations, and the risk that creditors eventually demand higher yields. On the other side, defenders argue a sovereign government borrowing in its own currency can always make nominal payments, the metric that matters is debt-to-GDP ratio rather than the raw number, and Japan has carried debt above 250% of GDP for years while still borrowing at rock-bottom rates. You do not need to pick a side for the AP exam, but you do need to understand both arguments.

Fiscal Policy and the Phillips Curve

Every AD shift from fiscal policy has a mirror on the Phillips Curve. Expansionary fiscal policy (more G or lower T) shifts AD right, real GDP climbs, unemployment falls, price level rises. Translate that to the short-run Phillips Curve and the economy slides up and to the left. Lower unemployment, higher inflation.

Contractionary fiscal policy traces the opposite path. AD shifts left, economy cools, and on the SRPC the economy moves down and to the right. Higher unemployment, lower inflation.

That trade-off holds in the short run. But there is a wall. If the government keeps injecting stimulus after the economy has returned to full employment, workers and firms start adjusting their inflation expectations upward and the entire short-run Phillips Curve shifts up. Unemployment drifts back to the natural rate eventually, but now with a permanently higher inflation baseline. The long-run Phillips Curve captures this. It is a vertical line at the natural rate of unemployment, same logic as the vertical LRAS at potential output. You can push unemployment below the natural rate temporarily with fiscal stimulus, but you cannot hold it there without accelerating inflation.

On the AP exam, an expansionary fiscal policy question should trigger three linked pictures in your head. AD/AS diagram where AD shifts right, output rises, price level rises. The money market where higher income raises money demand, pushing interest rates up. And the Phillips Curve where unemployment falls and inflation rises. Cleanly connecting all three models is where strong scores come from. That multi-graph linkage is basically the whole game on the free-response section.

Worked Example

Congress is debating two proposals to fight a recession. The MPC is 0.75.

Option A: spend $20 billion on infrastructure. Spending multiplier = 1 / (1 - 0.75) = 1 / 0.25 = 4. The full $20 billion enters the economy on day one. Construction workers spend 75% ($15 billion), those recipients spend 75% of that ($11.25 billion), and so on until the rounds converge. Total AD shift: $20 billion x 4 = $80 billion.

Option B: cut taxes by $20 billion. Tax multiplier = -0.75 / 0.25 = -3. Households get the $20 billion but save 25% upfront ($5 billion). Only $15 billion enters the first spending round. Chain: $15 billion, then $11.25 billion, then $8.44 billion, continuing until convergence. Total AD shift: $20 billion x 3 = $60 billion.

Spending wins by $20 billion in total AD impact. On an AD/AS graph, both shift AD right but the spending option shifts it further. The gap between the multipliers is always exactly 1 (4 vs. 3 here) regardless of MPC.

Both policies raise real GDP and the price level as the economy moves up along the SRAS curve. The practical question is just how much AD shift Congress wants per dollar of fiscal commitment.