Defense Spending as a Share of the Economy, From Korea to the 2010s

Is a bigger number always a bigger burden? Ask a person who earns ten times what he earned in 1980 whether his rent, which has also grown a great deal, is crushing him. He will tell you it depends what you compare it to. That is the entire argument of this post, and it is also the thing most budget commentary forgets to do.

Dollars without a denominator are noise. When someone tells you the Pentagon spends hundreds of billions a year, he has told you a large number and nothing else. The question that matters for the debt, the deficit and the size of government is how large defense is relative to the economy that pays for it and relative to everything else the government does. Let us do the arithmetic.

Defining the measures

Three terms first.

Outlays are what the federal government actually spends in a fiscal year. National defense is the budget category, called function 050 in the federal accounts, that covers the Department of Defense, the nuclear weapons work of the Department of Energy and a few smaller items. GDP, gross domestic product, is the value of everything the economy produces in a year.

Two ratios follow. Defense as a share of GDP tells you how much of the nation's output goes to defense. Defense as a share of federal outlays tells you how much of the government's budget it consumes. They answer different questions, and the claim that the Pentagon drives the debt depends on the second.

The source for everything below is the Historical Tables published by the Office of Management and Budget, which give outlays by function and as percentages of GDP and of total outlays back to 1940. The figures here are rounded and approximate, and a careful reader should check the tables for the exact values.

One: the Korean War peak, about 1953

In fiscal 1953, the last full year of the Korean War, defense outlays were about 14 percent of GDP. The federal government was spending a very large share of the economy's output on soldiers, aircraft, ships and ammunition, and defense accounted for roughly 70 percent of everything the government spent.

That figure deserves a pause. In the early 1950s, if you asked what the federal government did, the honest answer was that it ran a military. Medicare and Medicaid did not exist at all. The national budget was, to a first approximation, a defense budget.

Two: the Vietnam high, about 1968

By fiscal 1968, at the height of the Vietnam War, defense outlays were about 9 percent of GDP. Notice the direction: even in a major war, the share was well below the Korean peak. The economy had grown much faster than the military had, and the government had taken on new peacetime programs. Medicare and Medicaid had been signed into law in July 1965, and the Great Society programs were beginning to grow.

Defense was still the largest piece of the budget, at about 46 percent of outlays. But the share was already falling, and the reason was not that the military shrank. The rest of the budget grew.

Three: the Reagan high, about 1986

The Reagan buildup of the 1980s was a large peacetime increase in defense spending, and by fiscal 1986 defense outlays reached a little over 6 percent of GDP. That was a little over a quarter of federal outlays.

Pause on the contrast. At the height of what critics called a runaway military budget, defense took a smaller slice of the national economy than it had in the Vietnam year of 1968, and less than half the Korean peak. The buildup was large in dollars, and modest as a share of what the nation could afford.

That buildup also bought something. It helped bring about the end of the Cold War without a direct war between the superpowers, as explored in NSC-68 and the 1950 Memo That Told America the Price of Staying Free and Reagan at Westminster in 1982 Predicted the Soviet Collapse Out Loud.

Four: the late-1990s low

After the Cold War ended, defense spending fell. By the late 1990s, defense outlays were near 3 percent of GDP, among the lowest levels of the postwar record, and about 16 percent of federal outlays. The "peace dividend" had been collected.

The conservative lesson of that decade is that the dividend came due. The forces were smaller, the industrial base had consolidated, and the readiness of units declined. When the September 11 attacks came, and two long wars followed, the nation built forces back up at a premium price. Peace Dividends Always Come Due, the 1990s Drawdown Reconsidered tells that story in full.

Five: the 2010s

Defense rose in the 2000s because of the wars in Afghanistan and Iraq, peaking around 2010, and then fell back. By the end of the 2010s, before the pandemic, it was around 3.2 percent of GDP, near the low end of the postwar record, and roughly 15 percent of federal outlays.

Here are the rounded numbers in one place.

The trend is as clear as a trend can be. Over seventy years, defense fell as a share of the economy by roughly three-quarters, and as a share of the budget by about four-fifths.

What replaced it: the shift to mandatory programs

If defense fell as a share of everything, something else rose. The federal budget divides into discretionary spending, which Congress sets each year through appropriations, and mandatory spending, which is set by standing law and runs on autopilot unless Congress changes the formulas. Defense is the largest piece of discretionary spending. Social Security, Medicare and Medicaid are the largest mandatory programs.

Social Security began in 1935 and grew with the aging of the population and the expansion of benefits. Medicare and Medicaid started in 1965. Their costs rise with the number of beneficiaries and with the price of health care, not with any annual decision. By 2019 these programs, together with the other health care spending, accounted for something near half of federal outlays, while interest on the debt was a growing line of its own.

So the budget turned inside out. In 1953, defense was about 70 percent of the budget. By 2019, it was about a seventh, and the programs that run on autopilot were about half.

For a different angle on the same split between what Congress votes on and what runs on its own, see Follow the Defense Bill From Markup to Signing, a Citizen's Guide, which traces the part of the budget that does get an annual vote.

The arithmetic of the deficit

Now for the test that decides the question. Take the last full year before the pandemic, fiscal 2019, and use rounded figures, which you can confirm in the OMB tables and the Treasury's reports.

• Federal outlays were about 21 percent of GDP.
• Federal revenues were about 16.3 percent of GDP.
• The deficit was therefore about 4.6 percent of GDP. That is 21 minus 16.4, roughly, which gives 4.6.
• Defense outlays were about 3.2 percent of GDP.

Now subtract. 4.6 percent is the gap, and 3.2 percent is all of defense. If Congress had abolished the entire Department of Defense in 2019, closing every base, grounding every aircraft and dismissing every service member, which no one proposes, the deficit would still have been about 1.4 percent of GDP. The budget would not have balanced. This is not a close call.

Run it the other way. Suppose a reformer wanted to cut defense by a fifth, a deep cut by any standard. A fifth of 3.2 percent is 0.64 percent of GDP. Against a 4.6 percent deficit, that closes about 14 percent of the gap. One fifth of defense yields one-seventh of the deficit.

The honest conclusion is that the Pentagon is not why the debt grows. A reader can see how the total public debt keeps climbing on the national debt clock, where the figure was $40.56 trillion on November 5, 2026, up from $38.05 trillion a year earlier. That growth is driven by the combination of mandatory programs, interest and a revenue base that has not matched either, and the defense line is not the biggest part of it.

A household version of the same math

Picture a family, labelled as hypothetical, with a household income of 60,000 dollars. In year one it spends 8,000 on a security system, locks, insurance and a fence. In year thirty its income has grown to 150,000 dollars and it spends 12,000 on the same protections. The spending went up by half. The share of income went from about 13 percent to 8 percent. Which family is more burdened by its fence?

Now add the part the budget debate leaves out. Over the same thirty years, this family took on a mortgage, a pair of car loans and a large and growing medical bill that is set by a contract it signed and cannot easily change. When the monthly statement arrives and the family wonders why it is short each month, the fence is not the answer. Cutting the fence in half would save a few hundred dollars a month and leave the problem standing there in the kitchen.

That is the federal budget with the names changed. The fence is real, and a family should check that it paid a fair price and that the gate shuts. But a family that blames the fence for the shortfall will spend years arguing about the fence while the mortgage compounds. Washington has been having the fence argument for decades, and it has been a remarkably comfortable argument for everyone who would prefer not to discuss the contract.

The comfort is political, and it is worth saying so plainly. Defense is easy to criticize because it is visible, it is voted on every year, and it has contractors with names. The mandatory programs run on autopilot, they are loved by their beneficiaries, and touching them is hazardous to a political career. A reform movement that aims at the visible target and ignores the large one is practicing a kind of fiscal theater, and the audience pays for the tickets.

The strongest counterclaim

Here is the best case the other side can make, stated fairly. The critic says: percentages hide the real size. The United States spends more on defense in dollars than the next several countries combined, and the sum has a large absolute value. A budget that big contains waste, and waste is worth cutting regardless of what else is wrong. Further, the critic adds, you chose a flattering baseline. The share of GDP was lower in the 1990s partly because the economy was booming.

Both points have force. Let us answer them in order.

One. On absolute dollars, the comparison to other countries confuses a measure of spending with a measure of capability. America has global commitments, a volunteer force paid at market rates, and a navy that must cover two oceans. A country that spends less because it has a conscript army and a regional mission is not comparable. The question for a conservative is whether the spending buys the capability the mission requires, and that is a question about the mission and the management, not the total.

Two. On waste, we agree, and say so loudly. The Pentagon Has Never Passed an Audit, and Conservatives Should Mind argues that the Pentagon's financial management is indefensible, and Zumwalt, the Littoral Combat Ship and the Lessons of Procurement Failure shows programs that squandered billions. Waste in defense deserves the same hostility as waste anywhere. But a campaign against waste is a different campaign from a claim that defense is the engine of the debt. You can believe in a leaner Pentagon and still see that the arithmetic points elsewhere.

Three. On the baseline, the table above uses five different years across seventy years, and the trend is the same under any starting point. If you take the 2019 figure and compare it to the average of the postwar period, it is below the average, not above.

The Founding view

The Constitution treats defense differently from other spending, and the difference is instructive. Article I, Section 8 gives Congress the power "To raise and support Armies" but with a limit, that "no Appropriation of Money to that Use shall be for a longer Term than Two Years." The Founders wanted the army's money to come up for renewal every two years, so that the people's representatives would have to look at it regularly. Federalist No. 41 made the same point about the means of security, which "can only be regulated by the means and the danger of attack."

The principle fits our figures. Defense is the one function in the Constitution that has no natural ceiling, because the threats are set by others. That makes the share of the economy a more useful guide than the sum. The right question is whether the share matches the danger. Seventy years of declining share, at a time when great-power rivalry has not gone away, suggests that the nation has been getting its defense at a discount.

The case in brief

1. Measured against the economy, defense spending has fallen from about 14 percent of GDP at the Korean War peak, to about 9 percent in Vietnam, a little over 6 percent under Reagan and near 3 percent in the late 1990s and the late 2010s.
2. As a share of federal outlays, defense fell from about 70 percent in 1953 to about 15 percent by 2019, while Social Security, Medicare and Medicaid grew to about half.
3. The arithmetic of the 2019 deficit, 4.6 percent of GDP against defense at about 3.2 percent, shows that even eliminating defense would not have balanced the budget.
4. Waste in the Pentagon is real and deserves relentless attention, but the claim that the Pentagon drives the debt fails the arithmetic.
5. A conservative budget reform has to deal with mandatory programs and interest, or it is not a reform.

What you can do: open the OMB Historical Tables, find the percentage-of-GDP table for outlays by function, and check the five years in the table above for yourself; then send your representative one question about the mandatory side of the budget.