How Did America Forget How to Balance Its Budget?

I was born in September 1994, which means that some of my earliest years happened to coincide with one of the strangest fiscal periods in modern American history.

I obviously did not know this at the time. I was learning to read and riding a bicycle while the federal government was doing something it had almost forgotten how to do: bringing in more money than it spent. In fiscal year 1998, the United States recorded its first federal budget surplus since 1969. The surplus grew in 1999. Then, in fiscal year 2000, it reached $237 billion, or about 2.4 percent of the entire economy. Federal debt held by the public was actually being paid down.[1]

There was something almost futuristic about the mood surrounding the numbers. At the beginning of 2001, the Congressional Budget Office projected that, if the laws then on the books remained in place, the federal government would accumulate $5.6 trillion in surpluses between 2002 and 2011.

Instead, it accumulated $6.1 trillion in deficits.

That is an $11.7 trillion difference between the fiscal future the country could see from January 2001 and the one that actually arrived.[2]

I have been trying to understand what happened in between.

The obvious place to look is September 11. The attacks transformed American foreign policy, created an enormous new security apparatus, led directly to wars in Afghanistan and Iraq, and began commitments whose financial consequences are still arriving decades later. It is tempting to draw a clean line from the balanced budgets of my early childhood to the War on Terror and then onward to the immense public debt of the present.

There really is such a line.

It is just not the only one.

The surplus was more fragile than it looked

The first thing that complicates the story is that the fiscal world of 2000 was not simply the normal condition of the United States before September 11 knocked it off course.

It was exceptional.

The economy was unusually strong. Federal revenues were high. Interest rates were relatively favorable. The end of the Cold War had allowed defense spending to fall substantially as a share of the economy. Health-care spending was growing more slowly than expected. The demographic structure of Social Security was unusually favorable because the enormous baby-boom generation was still mostly working rather than collecting benefits.

In early 2002, the Government Accountability Office looked back at this combination and described the surplus years as something close to an “eye of the storm.” The country had benefited from a collection of economic, demographic and geopolitical tailwinds that were already beginning to disappear.[3]

Even before September 11, some of them had.

The technology bubble had burst. Economic growth was slowing. Tax receipts were weakening. And in June 2001, President George W. Bush signed a major package of tax cuts.

By August, before the attacks occurred, the budget picture had already deteriorated sharply. An analysis of contemporary CBO figures found that the year's newly enacted tax cuts reduced 2001 revenue by roughly \$74 billion, while the weakening economy further reduced collections.[4] The fiscal year would still end in surplus, at \$127 billion, but that was barely half the surplus of the year before.

So September 11 did not take a perfectly healthy budget and suddenly turn it into a deficit.

What it did was arrive at almost precisely the moment when an unusually favorable fiscal era was already ending.

Then it changed what the United States believed it had to spend money on.

The wars went on the credit card

The attacks created immediate costs, but the larger fiscal transformation came from the commitments that followed them.

The United States invaded Afghanistan in October 2001. Iraq followed in 2003. Military operations continued for years, alongside intelligence expansion, counterterrorism programs, airport security, border security, veterans' care and a permanent homeland-security bureaucracy.

The Department of Homeland Security was created in 2002. The Transportation Security Administration had been established the year before. Security procedures that would have seemed extraordinary in the summer of 2001 became ordinary pieces of American life.

There is no single official line in the federal budget called “the cost of September 11.” That makes accounting difficult. Spending is distributed across the Pentagon, Department of Homeland Security, Department of Veterans Affairs and other agencies, while some of the largest costs appear only years later as veterans age or the Treasury pays interest on money borrowed long ago.

Brown University's Costs of War project has attempted to assemble the pieces. Its broad estimate puts the eventual budgetary cost of the post-9/11 wars at roughly $8 trillion, counting military operations, increases in Pentagon base spending associated with the wars, homeland-security spending, veterans' care and interest on borrowed money. More than $1.1 trillion of its estimate comes from homeland-security spending associated with preventing and responding to terrorism, while another enormous share consists of present and future care for veterans.[5]

The exact figure can be debated because it depends on which indirect costs are attributed to the wars. The underlying point is harder to dispute.

The United States did not pay for the War on Terror as it went.

There was no large wartime tax introduced to cover Afghanistan and Iraq. Taxes had just been cut. Military operations were therefore financed largely through borrowing.

The cost of a patrol in Afghanistan in 2008 did not necessarily appear entirely on the tax bill of an American living in 2008. Part of it was converted into a Treasury security, rolled into the national debt and carried forward. Interest was added. Veterans returned home and required medical treatment and disability compensation. A war fought by one generation of soldiers became a fiscal obligation inherited by another generation of taxpayers.

In that sense, the wars never really ended when the troops came home.

Parts of them were securitized into the future.

But September 11 still cannot explain the whole thing

If I began this investigation wanting to know whether the country's inability to balance its budget was principally a consequence of September 11, this is where the evidence pushes back.

The attacks and their aftermath mattered enormously.

But America was making other expensive choices at the same time.

The Bush tax cuts reduced federal revenues. Congress also expanded Medicare in 2003 by creating the prescription-drug benefit that became Medicare Part D. The benefit addressed a genuine gap in health coverage for older Americans, but it was not accompanied by a dedicated source of revenue sufficient to pay for it. CBO initially estimated that the 2003 Medicare law would increase deficits by roughly $394 billion over its first decade, although Part D ultimately proved substantially cheaper than early projections.[6]

That combination is important.

The United States was simultaneously fighting new wars, constructing a new security state, lowering taxes and creating a major new entitlement benefit.

Any one of those choices can be defended.

The budget has no column labeled good intentions.

It only records what comes in and what goes out.

By 2001, debt held by the public had fallen to roughly 33 percent of GDP. That was the lowest level in decades. The country entered the century with considerable fiscal room to maneuver.[7]

Over the following years, it began using that room.

Then the floor collapsed.

The crash that made enormous deficits feel normal

The financial crisis of 2008 belongs in this story for a different reason.

Wars and tax cuts are political choices. A financial crisis is a shock. When millions of people lose jobs, businesses fail and incomes fall, a government can run a much larger deficit even without passing a single new spending program. Tax receipts collapse just as unemployment benefits and other automatic supports increase.

In 2009, federal revenues fell by 17 percent from the year before. Outlays jumped by 18 percent. The federal deficit reached roughly $1.4 trillion, or 9.9 percent of GDP, then the largest share of the economy since 1945. The recession itself caused much of the deterioration, while rescue programs, support for the financial system and the 2009 stimulus added to it.[8]

Running a deficit under those conditions was not inherently irresponsible.

This distinction matters.

A government is not a household, and balancing the federal budget every single fiscal year is not necessarily desirable. During a deep recession, war, pandemic or other emergency, borrowing can prevent an economic collapse or finance an urgent response that would be impossible to pay for immediately through taxes.

The stranger part of the American story is not that the government borrowed heavily in 2009.

It is that balanced budgets never returned after the emergency passed.

The economy eventually recovered. Unemployment fell. Asset prices rose. The financial system stabilized.

The deficit survived.

That suggests the central problem is not emergency borrowing itself. It is what happens when a political system learns to borrow during emergencies but loses the ability to produce surpluses during good times.

Economists have a name for versions of this tendency: deficit bias. Governments face enormous incentives to provide visible benefits today while postponing the less visible cost of paying for them. The benefits of a program or tax cut can be immediate and concentrated. The eventual cost of additional public debt is distributed among millions of people and across many future years.[9]

The emergency ends.

The political incentives do not.

The demographic bill arrives

Meanwhile, a much slower process was moving underneath the crises.

America was getting older.

The baby boomers who had contributed to the favorable fiscal arithmetic of the 1990s gradually became Social Security and Medicare beneficiaries. This was not a surprise. Demographers had known it was coming for decades. But knowing that a bill will arrive is not the same thing as saving enough to pay it.

Social Security and Medicare are not marginal pieces of the federal government. They are among its largest programs, and their costs rise automatically as more people qualify for benefits. Medicare faces the additional pressure of health-care costs.

This is one reason the story of the debt cannot sensibly be reduced to waste, foreign aid, congressional salaries or whatever comparatively tiny line item happens to attract public anger in a particular year.

The large structural forces are much less exotic.

They are retirement.

Health care.

Taxes.

Defense.

And, increasingly, interest.

By 2009, CBO was already warning that rising health-care costs and population aging would place the federal budget on an unsustainable long-run path unless revenues rose or projected spending fell.[10]

The warning was not difficult to understand.

It was difficult to act on.

Another tax cut, another ratchet

The same basic political arithmetic appeared again in 2017.

The Tax Cuts and Jobs Act reduced taxes on corporations and many individuals while making numerous other changes to the tax code. Supporters argued that lower tax rates would increase investment and economic growth. CBO later estimated that the law would indeed increase economic output relative to what otherwise would have occurred.

But the additional growth was not projected to make the tax cuts pay for themselves.

After accounting for macroeconomic feedback and added interest costs, CBO estimated that the law would increase federal deficits by roughly $1.85 trillion between 2018 and 2028.[11]

This happened during an economic expansion.

That detail matters.

Deficit spending during a crisis can be thought of as borrowing fiscal capacity from the future to stabilize the present. Deficit spending during normal economic conditions does something different. It reduces some of the room available when the next emergency arrives.

The next emergency arrived almost immediately.

Then the world stopped

COVID-19 produced exactly the sort of situation in which enormous government borrowing is easiest to defend.

Businesses were ordered closed. Tens of millions of workers suddenly faced the possibility of losing income. Hospitals confronted an unfamiliar disease. State and local governments saw revenues threatened while public-health demands exploded.

Washington responded on a scale that would have sounded almost impossible a few years earlier.

The major pandemic laws enacted in March and April 2020 alone were estimated by CBO to add about \$2.3 trillion to the fiscal 2020 deficit and another $600 billion in 2021. Additional relief followed in December 2020. Then the American Rescue Plan, enacted in March 2021, was estimated to add roughly another \$1.8 to \$1.9 trillion to deficits over the following decade.[12]

Some of that money replaced income that had vanished because society had deliberately suppressed economic activity. Some supported businesses, hospitals and governments. Some programs were undoubtedly better designed than others.

The important point here is not to relitigate every pandemic policy.

It is that another legitimate emergency arrived before the country had repaired the fiscal damage accumulated through previous ones.

In fiscal year 2020, the deficit reached $3.1 trillion.

America had discovered that a trillion-dollar deficit was possible during the financial crisis.

A decade later, it discovered that three trillion was possible too.

Eventually the debt sends its own invoice

There is a peculiar transition that happens when debt becomes large enough.

At first, borrowing finances things.

A war.

A tax cut.

A prescription-drug benefit.

A recession response.

A pandemic check.

A bridge.

A laboratory.

Eventually, some borrowing finances the cost of the borrowing itself.

This is where the federal budget of the 2020s begins to look meaningfully different from the federal budget of my childhood.

CBO's February 2026 projections put federal debt held by the public at roughly 101 percent of GDP this fiscal year, compared with 33 percent in 2001. It projects a 2026 deficit of about $1.9 trillion, even without a recession on anything resembling the scale of 2008 or 2020.[13]

And the interest bill is becoming enormous.

CBO projects approximately $1 trillion in net federal interest spending in 2026. By 2036, it projects about $2.1 trillion annually. At that point, interest would consume 4.6 percent of GDP and nearly equal all federal discretionary spending combined.[13]

Interest is unusual because it provides no new bridge, aircraft, pension check, medical treatment or research grant.

It is the price of decisions already made.

Debt has begun producing expenditures of its own.

So who benefits?

This was one of the questions that initially made me curious about the subject.

If the United States continues borrowing trillions of dollars, somebody must be receiving those dollars. Somebody owns the debt. Somebody earns interest. Somebody receives government contracts. Somebody gets the tax reduction or benefit whose cost was not immediately paid.

That intuition is correct.

But it becomes misleading if I turn it into a search for one group secretly responsible for the deficits.

There are too many beneficiaries.

Defense contractors are among the most visible. Brown University's Costs of War project and the Quincy Institute calculated that private companies received approximately $2.4 trillion in Pentagon contracts between 2020 and 2024, about 54 percent of Pentagon discretionary spending during those years. Five companies alone, Lockheed Martin, RTX, Boeing, General Dynamics and Northrop Grumman, received $771 billion.[14]

That creates obvious economic constituencies for high defense spending. Jobs, factories, shareholders, suppliers and congressional districts can all become connected to military procurement.

But defense contractors are only one piece.

Hospitals and pharmaceutical companies receive Medicare dollars.

Retirees receive Social Security.

Universities receive federal research money.

Farmers receive agricultural support.

States receive grants.

Businesses receive contracts.

Households receive tax credits.

And when Congress cuts taxes without reducing spending by the same amount, the beneficiaries are the people and companies who keep money that otherwise would have been collected.

A tax cut financed by borrowing is just as capable of creating a deficit as a spending increase financed by borrowing.

Then there are the lenders themselves.

Every Treasury security representing federal debt is simultaneously a liability of the United States government and an asset owned by somebody else. Treasury securities are held by households, mutual funds, pension funds, banks, corporations, the Federal Reserve and foreign investors. They also occupy an unusually important place in the global financial system because they are treated as exceptionally liquid and safe assets and are widely used as collateral.[15]

Their owners receive interest.

But this still does not mean Treasury investors are engineering federal deficits so that they can collect it. They are lending money the government has already decided to borrow. The existence of buyers makes persistent borrowing possible, but the buyers are not the same thing as the political decisions that create the need for borrowing.

The more interesting answer to who benefits? may therefore be:

Almost everyone benefits from some part of the system.

Just not from the whole thing.

The politics of the unpaid bill

That begins to explain why solving the problem is so difficult.

Imagine a budget assembled by millions of people who each get to circle the parts they like.

One person circles Social Security.

Another circles Medicare.

Another circles defense.

Another circles scientific research.

Another circles veterans' benefits.

Another circles infrastructure.

Another circles lower corporate taxes.

Another circles lower individual taxes.

Another circles disaster relief.

Another circles farm support.

Another circles housing assistance.

Almost nobody circles interest on the national debt.

Yet if we refuse to uncircle anything else, and refuse to collect enough money to pay for everything we circled, interest is what appears.

Political economists sometimes describe government revenue as a kind of common pool. The benefits of a particular expenditure or tax preference can be concentrated among the people receiving it, while the cost is spread across the entire taxpaying population and, when borrowing is involved, across people who may not even have been born yet.[16]

That creates a system capable of producing an outcome that almost nobody explicitly wants.

A voter can sincerely favor lower deficits while opposing cuts to the programs that matter to them.

A politician can sincerely favor lower deficits while opposing tax increases.

A business can favor fiscal responsibility while lobbying for a valuable federal contract.

A retiree can worry about the debt while depending on benefits financed partly by current workers.

A younger worker can resent entitlement costs while supporting tax reductions that also enlarge deficits.

There does not need to be a smoke-filled room in which someone decides to bankrupt the country.

The machine can run perfectly well on ordinary incentives.

Keep the programs.

Keep the tax cuts.

Fund the wars.

Respond to the emergencies.

And send some portion of the bill forward in time.

What September 11 actually changed

After following all of these threads, I would answer my original question differently than I expected.

September 11 did not cause America's entire modern debt problem.

But it was a major hinge.

It arrived just as the economic and demographic conditions supporting the late-1990s surpluses were beginning to weaken. It generated decades of direct military and security spending. It helped create permanent institutions and obligations. The wars that followed were largely financed through borrowing, allowing their cost to migrate into later budgets through interest and veterans' care.

Perhaps most importantly, September 11 was the beginning of a sequence in which extraordinary events kept arriving before the fiscal consequences of the previous ones had been resolved.

War.

Financial crisis.

Recession.

Pandemic.

Between them came tax cuts, new benefits, demographic aging and the ordinary expansion of commitments that are politically much easier to create than to withdraw.

The fiscal history of the twenty-first-century United States therefore looks less like one catastrophic decision than a geological formation.

Layer after layer.

Each one explicable.

Together, immense.

The country I thought was normal

There is something strange about realizing that the economic atmosphere of your childhood may have been the exception rather than the rule.

When I was born, the Cold War had just ended. By the time I was old enough to remember much, unemployment was low, technology companies were transforming daily life, federal revenues were booming, publicly held debt was falling and the United States occupied a position of extraordinary economic and geopolitical confidence.

That America had serious problems too. The late 1990s should not be converted into a political Eden simply because some numbers looked good.

But the fiscal trajectory was real.

In fiscal 2000, the government collected more than it spent.

In fiscal 2001, it did so again.

It has not done so since.

Twenty-five years later, CBO projects a federal deficit of roughly \$1.9 trillion for 2026 and debt held by the public larger than the country's annual economic output. By 2036, under current law, it projects debt at 120 percent of GDP and annual deficits exceeding \$3 trillion.[13]

That does not mean the United States is poor.

It remains astonishingly productive. It contains many of the world's most valuable companies, universities, laboratories, financial institutions and technology industries. It can still create extraordinary things.

But national wealth and government fiscal health are not the same measurement.

A country can remain rich while progressively narrowing its room for maneuver.

That may be the most important difference between the America at the beginning of my life and the one I inhabit now.

In 2000, the United States confronted the future with a federal balance sheet that seemed to offer choices.

A quarter century later, more and more of the budget is occupied by promises made in the past, interest on money already borrowed, and programs whose constituencies have reasonably organized their lives around expecting them to continue.

The tragedy, if there is one, is not that somebody stole all the money.

The story is more ordinary than that.

We spent some of it on things we needed.

We spent some on things we wanted.

We declined to collect some of the money required to pay for either.

Emergencies kept arriving.

The bills accumulated.

And because each individual bill could be explained, the growing stack somehow became easier to ignore.

Until eventually the stack itself became one of the largest bills.

Sources and further reading

1. U.S. Department of the Treasury, “Budget Results for Fiscal Year 2000.” Contemporary federal accounting of the \$237 billion FY2000 surplus, equal to 2.4 percent of GDP, and the accompanying reduction in debt held by the public.

2. Congressional Budget Office, “Changes in CBO's Baseline Projections Since January 2001.” A remarkable retrospective comparing the \$5.6 trillion in projected 2002–2011 surpluses with the \$6.1 trillion in deficits that actually occurred, an \$11.7 trillion swing from the January 2001 baseline.

3. U.S. Government Accountability Office, “Budget Issues: Long-Term Fiscal Challenges.” Published in 2002, this is especially valuable because it describes the extraordinary economic growth, slower health-care cost growth, favorable demographics and post-Cold-War defense dividend that helped produce the late-1990s surpluses, characterizing the period as an “eye of the storm.”

4. Center on Budget and Policy Priorities, “What Happened to the Surplus?” A contemporary August 2001 analysis of the rapidly deteriorating budget outlook before September 11, including the effects of the 2001 tax cuts and weakening economy.

5. Neta C. Crawford, “The U.S. Budgetary Costs of the Post-9/11 Wars.” Costs of War, Brown University, 2021. A broad accounting of military operations, homeland security, veterans' obligations and interest associated with the post-9/11 wars, including the project's approximately \$8 trillion estimate of total spending and obligations.

6. Congressional Budget Office, “A Detailed Description of CBO's Cost Estimate for the Medicare Prescription Drug Benefit.” 2004. CBO's original accounting showed that the 2003 Medicare law was expected to increase deficits by about \$394 billion over 2004–2013. See also CBO's later report, “Competition and the Cost of Medicare's Prescription Drug Program,” which explains why actual Part D spending subsequently came in substantially below the original projections.

7. U.S. Government Accountability Office, “Financial Audit: Bureau of the Public Debt's Fiscal Years 2001 and 2000 Schedules of Federal Debt.” Records the fall in debt held by the public from 43 percent of GDP in 1998 to approximately 33 percent in 2001, while already noting the fiscal pressures created by terrorism-related spending, tax-policy changes and deteriorating economic conditions.

8. Congressional Budget Office, “Federal Budget Deficit Totals \$1.4 Trillion in Fiscal Year 2009.” A contemporary account of the Great Recession's fiscal impact. Federal revenues fell by about 17 percent in 2009 while outlays increased by about 18 percent, producing a deficit equal to 9.9 percent of GDP.

9. Alberto Alesina and Roberto Perotti, “The Political Economy of Budget Deficits.” NBER Working Paper No. 4637, 1994; later published in IMF Staff Papers, 1995. doi:10.3386/w4637. A survey of political explanations for persistent government deficits, including fiscal illusion, intergenerational redistribution, geographically dispersed interests and the influence of budgetary institutions.

10. Congressional Budget Office, “Long-Term Budget Outlook.” 2009. CBO warned that rising health-care costs and population aging were placing the federal budget on an unsustainable long-run path and would require higher revenues, lower projected spending or some combination of the two.

11. Congressional Budget Office, “How the 2017 Tax Act Has Affected CBO's GDP and Budget Projections Since January 2017.” CBO's retrospective estimate found that, after accounting for projected economic feedback and debt-service effects, the Tax Cuts and Jobs Act would increase cumulative deficits by approximately \$1.85 trillion over 2018–2028.

12. Congressional Budget Office, “The Effects of Pandemic-Related Legislation on Output.” CBO estimated that the four major federal laws enacted in March and April 2020 would add about \$2.3 trillion to the FY2020 deficit and \$600 billion in FY2021. For the subsequent American Rescue Plan, see CBO's “Potential Statutory Pay-As-You-Go Effects of the American Rescue Plan Act of 2021,” which estimated an approximately \$1.9 trillion increase in deficits over 2021–2031.

13. Congressional Budget Office, “The Budget and Economic Outlook: 2026 to 2036.” The current baseline underlying several figures in this essay. CBO projects a \$1.9 trillion deficit and debt held by the public equal to 101 percent of GDP in 2026, rising to a \$3.1 trillion deficit and 120 percent of GDP in debt by 2036. Net interest costs are projected to rise from about \$1.0 trillion in 2026 to \$2.1 trillion in 2036.

14. William D. Hartung and Stephen N. Semler, “Profits of War: Top Beneficiaries of Pentagon Spending, 2020–2024.” Costs of War / Quincy Institute, 2025. The authors calculate that private firms received approximately \$2.4 trillion in Pentagon contracts over those five years, including \$771 billion received by five major weapons manufacturers.

15. U.S. Department of the Treasury, “Holders of U.S. Treasuries Over Time.” Treasury Borrowing Advisory Committee materials showing the broad range of Treasury holders, including foreign investors, the Federal Reserve, households, money-market funds, asset managers, banks, pensions and insurers. For the special financial role of Treasury securities, see the Federal Reserve Bank of New York's “Dollar Asset Markets: Prospects after the Crisis,” which discusses their unusually large, liquid and safe-haven market.

16. Jürgen von Hagen and Ian Harden, “Budget Processes and Commitment to Fiscal Discipline.” IMF Working Paper, 1996. A classic treatment of the “common pool” problem in government budgeting: the benefits of particular expenditures can be concentrated among specific groups while their financing costs are spread much more widely, creating incentives toward excessive spending and debt.