Fiscal policy, taxes and public debt
Governments spend, tax and borrow on a scale that moves markets, and their debt sets the base for every other interest rate. This lesson separates the deficit, the debt and the interest bill, shows how automatic stabilizers work, and explains why a debt ratio rises or falls. A calculator starts from the US debt held by the public and shows how the interest rate, growth and the budget move the ratio. It compares the US with Japan on the IMF's measure, and ends with the effective tax rates of Alphabet and Walmart.
- Level
- Beginner
- Reading time
- About 29 minutes
- Know this first
- Exchange rates and payments
By the end you will be able to:
- tell a deficit, a primary deficit and a debt apart
- explain discretionary policy and automatic stabilizers
- use the debt-ratio rule to see how the interest rate, growth and the primary budget interact
- compare countries' debt on one definition and say what the ratio does not show
- calculate an effective tax rate and say how it differs from the statutory rate
The government as a very large borrower
The last lessons were about central banks and their interest rates. This one is about the other half of macroeconomic policy: the government’s own budget. Fiscal policy is the use of taxes and public spending to influence the economy. It matters to finance for three reasons. Government bonds are the safest and most traded assets in the world, so the yield curve of the previous lesson is really the price of government borrowing. Taxes take a part of every company’s profit. And the size of a government’s debt, and the interest it pays, can limit what it does later.
The lesson uses US government figures from the Treasury and the Office of Management and Budget, and then does something unusual for a budget lesson: it gives you a calculator to see how a debt ratio moves from year to year. It also compares the US and Japan, and looks at how much tax two companies actually paid.
Deficits, debt and interest
Three words are often confused. A deficit is the gap in one year when the government spends more than it collects in taxes and other receipts. Debt is the stock built up from past deficits. And interest is the yearly cost of carrying that stock. The table uses fiscal years, which for the US government end on 30 September. Each year’s outlays are receipts plus the deficit.
| Year ending 30 September | Receipts, $ billions | Outlays, $ billions | Deficit, $ billions | Interest outlays, $ billions | Interest as share of receipts |
|---|---|---|---|---|---|
| Fiscal 2019 | 3,463.4 | 4,447.0 | −983.6 | 375.2 | 10.8% |
| Fiscal 2020 | 3,421.2 | 6,516.8 | −3,095.7 | 345.5 | 10.1% |
| Fiscal 2021 | 4,047.1 | 6,820.7 | −2,773.6 | 352.3 | 8.7% |
| Fiscal 2023 | 4,440.9 | 6,128.4 | −1,687.5 | 658.3 | 14.8% |
| Fiscal 2024 | 4,919.9 | 6,735.3 | −1,815.4 | 879.9 | 17.9% |
| Fiscal 2025 | 5,236.4 | 7,011.1 | −1,774.7 | 970.1 | 18.5% |
Look at the shape. In fiscal 2019 the deficit was $983.6 billion. In 2020 it jumped to $3,095.7 billion, as outlays rose 46.5% in one year while receipts barely changed. By 2025 the deficit had settled at $1,774.7 billion, with receipts at $5,236.4 billion and outlays at $7,011.1 billion. The Congressional Budget Office (CBO) puts the fiscal 2025 deficit at 5.9 percent of GDP, against a 50-year average of 3.8 percent, and notes that its own measure of net interest passed $1 trillion for the first time. CBO’s measure differs from the series in the table, which shows interest outlays of $970.1 billion, so the two should not be mixed. The last column is the one to watch. Interest took 10.8% of receipts in 2019 and 8.7% in 2021. By 2025 it took 18.5%, because the debt had grown and the rates paid on it had risen.
Fiscal policy and the business cycle
Fiscal policy works in two ways. Discretionary policy is when the government passes a new law that explicitly changes tax or spending levels. OpenStax gives the 2020 stimulus checks as an example. Automatic stabilizers are rules already in law that change taxes and spending without a new vote. When the economy weakens, incomes and profits fall, so tax payments fall, while unemployment rises, so benefit payments rise. Both push the budget toward more support for demand. In an expansion the reverse happens.
The stabilisers are real but limited. OpenStax says they “do not offset all or even most” of a shift in demand, and that historically they have offset about 10% of an initial change in output. Their strength depends on the size of government: they were much weaker around 1900 and 1929, when federal spending was a small share of GDP, than in recent decades. This links directly to the earlier GDP lesson. The jump in the deficit in 2020, in the table above, is what a recession does to a budget, through lower receipts, automatic spending and new laws together. The table does not separate the three.
Why a debt ratio moves
A government’s debt is usually measured against the size of its economy, as a percentage of GDP. That ratio changes for two reasons. It rises when the government borrows to cover a primary deficit, the deficit before paying interest. And it rises when interest accumulates on the existing debt. But it falls when the economy grows, because GDP, the bottom of the ratio, gets bigger. Putting these together gives a one-year rule.
The key comparison is between the interest rate and the growth rate of nominal GDP. When growth is higher than the rate, the debt ratio shrinks by itself, and the government can run a primary deficit and still hold the ratio steady. When the rate is higher than growth, the ratio climbs unless the government runs a primary surplus. The same logic applies to a company whose debt grows by more than its profits, or to a household whose mortgage costs more than its income rises. The debt held by the public is the figure used here, which is the part of federal debt owed outside the government itself.
The series for that ratio, from OMB, is below, with the dollar amounts from the Treasury Bulletin. It stood at 34.9% of GDP at the end of 2007, 78.4% at the end of 2019, 96.9% by the third quarter of 2020, and 98.7% in the first quarter of 2026.
| Quarter end | Debt held by the public, $ trillions | Percent of GDP |
|---|---|---|
| Q4 2007 | 5.14 | 34.9% |
| Q4 2019 | 17.19 | 78.4% |
| Q3 2020 | 21.04 | 96.9% |
| Q4 2023 | 26.96 | 94.8% |
| Q4 2025 | 30.87 | 98.2% |
| Q1 2026 | 31.45 | 98.7% |
Move the debt ratio
Start from the first quarter of 2026 figure of 98.7% of GDP. Set the interest rate the government pays on its debt, the growth of nominal GDP and the primary deficit. The calculator moves the ratio one year ahead and shows what interest costs. It starts with figures from the recent record: an interest rate of 3.3%, which is fiscal 2025 interest outlays divided by the average debt held by the public over that year, nominal GDP growth of 6.3%, which is the rise from the second quarter of 2025 to the second quarter of 2026, and a primary deficit of 2.6% of GDP, which is the fiscal 2025 deficit less interest, divided by GDP. Revenue is taken as 16.8% of GDP, the fiscal 2025 figure.
With an interest rate of 3.30% on the debt, nominal GDP growth of 6.30% and a primary deficit of 2.60% of GDP, debt held by the public moves from 98.71% to 98.52% of GDP, a fall of 0.19 points. Interest costs 3.26% of GDP, or 19.40% of federal revenue. A primary deficit of up to 2.79% of GDP would hold the ratio steady.
Four experiments. First, leave the starting settings alone. Growth is well above the interest rate, so the ratio barely moves, a change of −0.19 points, even with a primary deficit of 2.6% of GDP, and the deficit that would hold it steady is 2.79%. Second, set the interest rate to 5% and growth to 4%. Now the rate is above growth. The ratio rises 3.55 points in one year to 102.26%, interest takes 29.39% of revenue, and holding it steady would need a primary surplus of 0.95%. Third, set both to 3% and the primary budget to balanced. The ratio does not move at all. When the rate equals growth, a balanced primary budget holds the ratio steady. Fourth, return to 5% and 3%, and set a primary surplus of 1%. The ratio still rises 0.92 points, because the interest on a debt of almost 99% of GDP outruns the surplus. The size of the debt multiplies the gap between the rate and growth.
Two countries, one measure
The International Monetary Fund (IMF) publishes general government gross debt as a percentage of GDP for each country on the same definition. It covers the whole government, including state and local, so it is larger than the federal figure from OMB used above. Japan and the United States make a good pair, because they are both large, rich and heavily indebted, and yet their debt ratios differ by a factor of two.
| Year | Japan | United States |
|---|---|---|
| 2007 | 173.0% | 64.7% |
| 2010 | 205.9% | 95.3% |
| 2015 | 228.3% | 104.9% |
| 2019 | 236.4% | 108.2% |
| 2020 | 258.4% | 132.0% |
| 2022 | 248.3% | 118.8% |
| 2023 | 240.0% | 119.0% |
In 2023, Japan’s gross debt was 240.0% of GDP and the United States’ was 119.0%. Both rose sharply in the pandemic year of 2020 and then eased. Japan’s ratio has been above 170% since the start of this table. A larger ratio does not by itself mean a larger burden: the burden depends on the interest rate and growth, as the calculator shows. The IMF series for Japan used here ends in 2023, so the table stops there. The sources used do not give Japan’s interest costs, so the comparison cannot say which country carries the heavier burden.
Taxes and the company
Taxes are the other side of the budget, and every company pays them. The IRS describes the Tax Cuts and Jobs Act’s flat 21 percent rate on corporate income, which generally applies to tax years beginning after 31 December 2017. But a company’s effective tax rate, its income tax expense divided by its pre-tax income, is usually not the statutory rate. State taxes, income earned in other countries, credits and deductions all move it. Neither company below states the rate in the release read, so it is calculated here from the income statement.
| Alphabet, calendar year | Walmart, year to 31 January | |
|---|---|---|
| Earlier year: pre-tax income, tax provision, effective rate | 2024: $119,815 million, $19,697 million, 16.4% | Fiscal 2025: $26,309 million, $6,152 million, 23.4% |
| Latest year: pre-tax income, tax provision, effective rate | 2025: $158,826 million, $26,656 million, 16.8% | Fiscal 2026: $29,469 million, $7,199 million, 24.4% |
| Compared with the 21% statutory federal rate | Below | Above |
Alphabet, the owner of Google, earned $158.8 billion before tax in 2025 and recorded a tax provision of $26.7 billion, an effective rate of 16.78%. Walmart, the retailer, earned $29.5 billion before tax in the year to 31 January 2026 and provided $7.2 billion, an effective rate of 24.43%. One rate is below the 21% statutory rate and the other is above it. The releases read here do not explain either rate, so what drives the difference is not shown. Walmart guides to a rate of approximately 23.5% to 24.5% for its next fiscal year. Rates also move from year to year, as both companies’ two columns show. For an analyst, the effective rate is a forecasting assumption: profit after tax is profit before tax times one minus the rate, and a small change in the rate moves earnings. For example, at Walmart a one-point fall in the rate would raise net income by about 295 million dollars on the fiscal 2026 pre-tax income.
What this lesson leaves out
- One country’s budget in depth. The US figures come from OMB and the Treasury. Other governments report on different definitions and calendars.
- The debt rule is an identity, not a forecast. The calculator assumes one interest rate, one growth rate and one deficit for the whole year. Real debt also changes with items outside the deficit, such as changes in cash balances and lending programmes, so it will not match the published change exactly.
- The starting settings are averages. The 3.3% is interest outlays over the average debt, not the rate on any particular bond. The path of rates on new borrowing is a different matter and was the subject of the previous lessons.
- Measures of interest differ. The OMB series, the CBO’s net interest and the debt measures all use somewhat different definitions. The lesson says which is used and avoids mixing them.
- No view on what a government should do. The lesson shows arithmetic and records. Whether to run deficits is a political and economic debate with serious arguments on several sides.
- Two companies, one year. They show that effective rates differ. They are not a sample, and the sources used do not explain the difference.
Common mistakes
Mixing up the deficit and the debt
The deficit is one year’s gap. The debt is the total built up over many years.
Looking at the debt without the economy
A debt of $31 trillion means little without GDP, the interest rate and growth. Ratios and rates are what matter.
Assuming a higher debt ratio always means a bigger burden
The burden depends on interest costs relative to income. A country can carry a higher ratio at a lower rate.
Treating the statutory rate as what companies pay
The effective rate is the one that reaches the income statement, and it varies by company and year.
Comparing debt figures from different sources
Debt held by the public, total public debt and general government gross debt are different measures. Compare like with like.
Try it yourself
1. In fiscal 2025 federal receipts were $5,236.421 billion and outlays were $7,011.105 billion. What was the deficit?
Show the working
5,236.421 − 7,011.105 = −1,774.684, a deficit of $1,774.7 billion.
2. Interest outlays in 2025 were $970.065 billion. What share of receipts is that?
Show the working
970.065 ÷ 5,236.421 = 18.5%.
3. The debt ratio is 98.71%, the interest rate is 5%, nominal growth is 4% and the primary deficit is 2.6% of GDP. What is the ratio a year later?
Show the working
98.71 × 1.05 ÷ 1.04 + 2.6 = 102.26%, a rise of 3.55 points.
4. With a rate of 3.3% and growth of 6.3%, what primary deficit holds the ratio steady?
Show the working
98.71 × (6.3 − 3.3) ÷ (100 + 6.3) = 2.79% of GDP.
5. Alphabet’s 2025 income before tax was $158,826 million and its tax provision was $26,656 million. What was the effective rate, and how does it compare with 21%?
Show the working
26,656 ÷ 158,826 = 16.78%, which is below the 21% statutory rate.
Where this is used
- A bond investor watches deficits and debt because government borrowing needs help set the yields of the curve.
- An analyst forecasting a company’s earnings applies an effective tax rate to pre-tax profit.
- A central bank considers the budget when it forecasts growth and inflation.
- A ratings agency or lender judges a government’s debt by its ratio to GDP, its interest costs and its growth.
- A business planning for a downturn thinks about which taxes and benefits will cushion demand.
Key terms
- Fiscal policy
- Using government taxes and spending to influence the economy.
- Deficit
- The shortfall in one year when spending exceeds receipts.
- Public debt
- The total a government owes from past borrowing.
- Primary deficit
- The deficit before interest payments.
- Debt held by the public
- Federal debt owed to holders outside the government itself.
- Automatic stabilizers
- Taxes and benefits that change with the economy without new legislation.
- Discretionary fiscal policy
- A new law that explicitly changes taxes or spending.
- Statutory tax rate
- The rate set in law.
- Effective tax rate
- Income tax expense divided by pre-tax income.
- Fiscal year
- A twelve-month accounting period, which for the US government ends on 30 September.
Questions people ask
Is a government deficit always bad?
Not by itself. Deficits widen automatically in recessions and can support demand. The concern is whether debt and interest costs grow faster than the economy.
Why does the interest rate matter so much for debt?
Interest is added to the debt each year. If the rate is above the growth of GDP, the ratio rises unless the government runs a primary surplus.
Why do company tax rates differ from the statutory rate?
Because of state taxes, income earned abroad, credits and deductions. The releases read for this lesson do not break down the reasons for these two companies.
Why does the lesson use debt held by the public?
Because it excludes debt the government owes to its own accounts, so it tracks borrowing from outside holders and fits the debt rule better than total public debt does.
Sources and review
- US Office of Management and Budget, Federal receipts (FYFR), millions of dollars, annual by fiscal year, as published by the Federal Reserve Bank of St. Louis (last updated 3 April 2026), for receipts.
- US Office of Management and Budget, Federal net outlays (FYONET), millions of dollars, annual by fiscal year, as published by the Federal Reserve Bank of St. Louis (last updated 3 April 2026), for outlays.
- US Office of Management and Budget, Federal outlays: interest (FYOINT), millions of dollars, annual by fiscal year, as published by the Federal Reserve Bank of St. Louis (last updated 3 April 2026), for interest outlays.
- US Office of Management and Budget, Federal surplus or deficit (FYFSD), millions of dollars, annual by fiscal year, as published by the Federal Reserve Bank of St. Louis (last updated 3 April 2026), for the deficits.
- US Department of the Treasury, Fiscal Service, Federal debt held by the public (FYGFDPUN), millions of dollars, quarterly, end of period (Treasury Bulletin; last updated 9 October 2026), for the dollar amounts and the average debt used for the interest rate.
- US Office of Management and Budget, Federal debt held by the public as percent of gross domestic product (FYGFGDQ188S), quarterly, as published by the Federal Reserve Bank of St. Louis (last updated 25 June 2026), for the debt ratios.
- International Monetary Fund, General government gross debt for Japan (GGGDTAJPA188N), percent of GDP, annual (World Economic Outlook; as published by the Federal Reserve Bank of St. Louis, last updated 29 April 2025).
- International Monetary Fund, General government gross debt for the United States (GGGDTAUSA188N), percent of GDP, annual (as published by the Federal Reserve Bank of St. Louis, last updated 29 April 2025).
- Congressional Budget Office, Monthly Budget Review: Summary for Fiscal Year 2025 (10 November 2025), for the deficit as a share of GDP, the 50-year average and net interest.
- US Bureau of Economic Analysis, Gross domestic product (GDP), billions of dollars, seasonally adjusted annual rate, quarterly (as published by the Federal Reserve Bank of St. Louis, last updated 30 September 2026), for nominal GDP growth and the GDP used for shares.
- OpenStax, Automatic stabilizers (Principles of Economics 3e, section 30.5), for the definitions of discretionary policy and automatic stabilizers and their limits.
- Internal Revenue Service, Tax Cuts and Jobs Act: businesses, for the flat 21 percent corporate rate.
- Alphabet Inc., Fourth quarter and fiscal year 2025 earnings release, for income before income taxes and the provision for income taxes.
- Walmart Inc., Fourth quarter and fiscal 2026 earnings release (Form 8-K), for income before income taxes, the provision for income taxes and the guidance for the effective tax rate.
How the numbers were checked: the deficits, the shares of receipts, the debt-ratio arithmetic, the effective tax rates and the sensitivity were recomputed in plain code from the figures in the sources above. Each year’s outlays in the table were checked as receipts plus the deficit against the published outlays series. The calculator, which runs separately, was compared with that code for every combination of settings it allows. Starting values for the calculator are derived as described in the text and are rounded.