Fiscal Policy

Expansionary and Contractionary Fiscal Policy

Fiscal Policy is how Congress uses spending and taxes to influence overall demand, from huge stimulus packages to the automatic ways the system stabilizes itself.

Fiscal policy is simply Congress using its ability to tax and spend to change overall demand. That's the entire idea in one sentence.

The CARES Act from March 2020 is a really clear illustration of this in the real world. Congress approved $2.2 trillion in spending, sending $1,200 to approximately 160 million individuals, providing $25 billion for airline payrolls, $175 billion for hospitals, and around $800 billion to small businesses in two phases of the Paycheck Protection Program. March and April saw 22 million jobs lost and the economy (GDP) shrank at a 31.4% rate in the second quarter. Because people had drastically cut back on buying things, the government took over as the main buyer, and this is the most basic form of expansionary fiscal policy. Increasing government spending (G), lowering taxes (T), or doing both moves the Aggregate Demand (AD) curve to the right. Consequently, Real GDP increases and prices also go up. The 2009 American Recovery and Reinvestment Act did the same thing during the Great Recession, using $831 billion on things like improving infrastructure, helping state governments financially, and giving tax breaks to the middle class.

Contractionary fiscal policy is the opposite. When the economy is producing more than it should and there's an inflationary gap, the theory says you should reduce spending or increase taxes to shift AD to the left. It's simple enough to write down. However, it almost never happens in practice; politicians who vote for higher taxes or cuts to things people like usually lose their jobs! So you need to thoroughly understand the theory for the AP exam, but also realize that actually doing contractionary fiscal policy is extremely rare in American politics.

The Spending Multiplier

When the federal government gave $42 million to rebuild the I-35W bridge in Minneapolis following its collapse in 2007, that money didn't just stay in the bank account of the construction company. Flatiron-Manson hired welders, crane operators, and engineers. These people then deposited their pay and used some of it at grocery stores, restaurants, and car dealerships in and around Minneapolis and St Paul. The grocery stores then used the money they received to pay their own employees, who in turn spent some of it. Each subsequent round of spending was a little less, because a portion of every dollar is saved instead of spent. This continuing effect is the multiplier effect.

The calculation is easy: Spending multiplier = 1 / (1 - MPC). The marginal propensity to consume (MPC) is the amount of each additional dollar people spend as opposed to save. An MPC of 0.8 means people spend 80 cents and save 20 cents of every new dollar. Plugging that in: 1 / (1 - 0.8) = 1 / 0.2 = 5. Therefore, a $10 billion building project would move AD to the right by $50 billion as that original spending goes through round after round. If the MPC went up to 0.9, the multiplier would be 10, and each dollar would pass through more people's hands before being put into savings. A higher MPC means a larger multiplier and a more effective stimulus for each dollar the government spends.

The Tax Multiplier

This is a frequent question on the AP Macro exam. Why doesn't a $10 billion tax cut have as much effect as $10 billion in direct government spending?

Consider what happens in the first phase. When the government spends $10 billion on building roads, all $10 billion immediately enters the economy because the government is actively buying something from someone. A $10 billion tax cut goes into people's bank accounts, but if the MPC is 0.8, households will save $2 billion of it right away. Only $8 billion actually gets spent in the first round. The chain of spending starts at a lower point and remains smaller.

The formula is Tax multiplier = -MPC / (1 - MPC). The negative sign is because taxes and AD go in opposite directions: a tax cut increases AD, and a tax increase decreases it. Using an MPC of 0.8: -0.8 / 0.2 = -4. That $10 billion tax cut shifts AD to the right by $40 billion. In comparison, $50 billion is the amount that would be created with direct government spending.

The difference in size between how much effect government spending has versus a tax cut is always exactly one, no matter the Marginal Propensity to Consume (MPC). This is because of that initial amount of money that isn't spent. The government spends all of the money immediately, but a tax cut loses whatever portion of it people save before anyone even spends anything.

Crowding Out

When figuring the multiplier effect, we usually assume that the government's spending just adds to the total amount of demand, and doesn't cause any other issues. But stimulus which is funded by borrowing has a problem.

When the government spends more than it takes in, the Treasury Department sells bonds to make up the difference. These bonds are competing with businesses (for expanding factories), developers (for building houses), and families (for car loans) for available money to borrow. More borrowers and a limited amount of savings available to lend causes interest rates to go up. A factory that would have been built if the interest rate was 4% is now put off with a 7% rate. A developer won't build those apartments. A family decides they can't afford the monthly car payments. Because of this, private investment declines, reducing the positive impact of the government spending. This is called the crowding-out effect, and it means the increase in Aggregate Demand (AD) isn't as much as a standard multiplier calculation would predict.

How much crowding out happens depends on the situation. During the 2008 financial crisis, private investment was already down significantly and interest rates were almost zero. There wasn't much to be crowded out, so the government spending wasn't stopped by much. But contrast that with an economy operating at full capacity where businesses are busily trying to get hold of every dollar of savings. In that case, government borrowing can push rates up quite a lot and significantly reduce private investment. In theory, "full" crowding out is possible, where for every dollar the government spends, a dollar of private investment is lost, and there is no overall increase in AD. But in reality, partial crowding out is more typical.

Automatic Stabilizers

Congress takes time. Hearings in committees, debates on the floor of the House and Senate, negotiations between the House and Senate, the President's signature. Even the 2009 Recovery Act, considered fairly quick, took nearly two months from the Inauguration to being signed into law. By the time the money finally gets into the economy, the recession might have gotten worse, or, rather strangely, even be over. Automatic stabilizers avoid all of this because they start working without a need for new laws.

The most important of these is the progressive income tax. If a worker loses their job or their hours are reduced, their income for tax purposes goes down and they automatically move into a lower tax bracket. Their spending money doesn't fall as much as their total income did. And during good economic times, as incomes increase, people are pushed into higher tax brackets, which takes purchasing power out of the economy without needing a vote in Congress. It's a natural way to slow down AD.

Unemployment insurance works in the opposite direction. When the economy shrinks, more people become eligible for benefits, and no new law is needed. The number of claims jumped from 211,000 a week in February 2020 to 6.9 million a week by late March, and most of that money went to people who spent all of it, helping to maintain consumer spending. Enrollment in SNAP (food stamps) works on the same principle, increasing automatically as incomes fall.

They reduce the size of ups and downs. They won't end a serious recession on their own. They're more like shock absorbers, not the engine.

Deficits and the National Debt

In the fiscal year 2020 the federal government spent approximately $6.55 trillion and collected around $3.42 trillion in taxes. That $3.13 trillion difference is the budget deficit, the largest in US history in actual dollar terms, although deficits during World War II (1943-1945) were a larger percentage of GDP. The Treasury Department covered this shortage by selling bonds, notes and bills to pension funds, central banks of other countries, individual savers, commercial banks, and even the Federal Reserve itself.

A deficit is about a single year - how much the government is 'in the red' for that time. The national debt, however, is the total amount of money the government owes, all the outstanding 'IOUs' at any given time. You get the national debt by adding up all the deficits that have ever happened, and then taking away any years where the government actually took in more money than it spent (the most recent of those was from and including 1998 to 2001).

Ideally, the budget should balance out over the course of a whole period of economic expansion and contraction. We'd have deficits during downturns (recessions) to increase spending and keep things going, and surpluses when the economy is doing well (booms) to save for the future. In reality, surpluses are extremely unusual. Reducing government spending or raising taxes is politically difficult no matter how the economy is performing. Even during the long period of growth from 2010-2019, the US still had deficits.

There's a lot of argument among economists about whether a large debt is a real threat to the economy. Those who are worried about it point to the increasing amount of the federal budget going to pay interest on the debt, the possibility of future tax increases to cover these payments, and the chance that people who lend to the government will eventually want a higher return for doing so. On the other hand, people who say it's not a huge problem say a country's government can always make payments in its own money. They argue that the debt-to-GDP ratio (how big the debt is compared to the size of the economy) is more important than the actual number of the debt, and Japan has managed to borrow more than 250% of its GDP for years and still get very low interest rates. You don't have to choose a side for the AP exam, but you do need to understand both viewpoints.

Connecting AD/AS, Money Market, and Phillips Curve

Every change in total demand (AD) caused by government spending or taxes is reflected on the Phillips Curve. When the government increases spending or cuts taxes (expansionary fiscal policy), AD shifts to the right. Real GDP increases, unemployment goes down, and prices go up. On the short-run Phillips Curve, this shows as the economy moving up and to the left: less unemployment, more inflation.

A reduction in government spending or tax increases (contractionary fiscal policy) follows the opposite pattern. AD shifts to the left, the economy slows down, and on the SRPC the economy moves down and to the right. This means more unemployment, and lower inflation.

This trade-off is true in the short term. However, there's a limit. If the government continues to stimulate the economy after it has already reached full employment, people and businesses will begin to expect higher inflation and the entire short-run Phillips Curve will move up. Unemployment will eventually go back to its 'natural' level, but now inflation will be permanently higher. The long-run Phillips Curve shows this, being a vertical line at the natural rate of unemployment (much like the vertical long-run aggregate supply curve at potential output). You can temporarily lower unemployment below the natural rate by increasing government spending, but you can't keep it there without inflation speeding up.

When you get a question about expansionary fiscal policy on the AP exam, you should think of three diagrams. First, the AD/AS diagram showing AD shifting to the right, with output and price levels increasing. Second, the money market, where the increased income increases the demand for money and pushes interest rates up. And third, the Phillips Curve, with unemployment falling and inflation increasing. Being able to clearly link all three of these models is the key to getting a good score - that connection between graphs is really what the free-response section is all about.

Worked Example

Here's an example: Congress is considering two ways to help with a recession, and the marginal propensity to consume (MPC) is 0.75.

Option A: spend $20 billion on building things.
The spending multiplier is 1 / (1 - 0.75) = 1 / 0.25 = 4. The full $20 billion is in the economy right away. Construction workers spend $15 billion (75% of it), those who receive that money spend 75% of that ($11.25 billion), and this continues as the money goes around. Eventually, the total increase in AD is $20 billion x 4 = $80 billion.

Option B: cut taxes by $20 billion.
The tax multiplier is -0.75 / 0.25 = -3. People get the $20 billion, but save $5 billion (25%). So, only $15 billion is spent in the first round. This continues with $15 billion, then $11.25 billion, then $8.44 billion, and so on. The total increase in AD is $20 billion x 3 = $60 billion.

Direct spending will increase overall demand by twenty billion dollars more than the tax cut. On an aggregate demand/aggregate supply graph, both will move the aggregate demand curve to the right, but the spending option will move it further. And the difference in how much each policy increases things (the multipliers) will always be exactly one (in this case, 4 versus 3), no matter the marginal propensity to consume. Both policies increase both real GDP and prices as the short-run aggregate supply curve is travelled up. Essentially, Congress needs to decide how much aggregate demand to boost for every dollar they spend.