The Bombing Windows Fallacy: How excessive defense spending hurts the U.S. economy

By Andrew Doris


Image: David Saveliev

Readers of Realist Review will be well acquainted with the moral and pragmatic case against military adventurism.  The external impact of an overactive US military is tragic and strategically damaging; in light of this, debating its internal impact seems almost heartless, and beside the point.  The United States should withdraw from its failed military interventions abroad, and avoid such entanglements moving forward, whether they help its economy or not.

Even so, it’s worth noting that maintaining the world’s most expensive military hurts Americans during peacetime, too.  Only about 10% of the $700 billion guzzled by the Defense Department each year is for “Overseas Contingency Operations” like the War on Terror; even after we bring the troops home, we’ll have still have to contend with a baseline defense budget of $629 billion.


And this does not include “defense-related” spending like Veterans Affairs ($186 billion) or the Dept. of Homeland Security ($44 billion).  All told, the United States spends nearly a trillion dollars a year on defense-related expenditures – and much of it would be better spent on something else.

Economic advocates of such colossal spending cite its positive effect on employment, innovation, and GDP.  The Department of Defense is the nation’s largest employer, making free healthcare, college education, and upward income mobility accessible to millions of people (and especially young people).  These jobs are largely “shovel ready,” making them easy to accelerate for short-term stimulus.  And over the long term, the economy has benefitted from technological breakthroughs pioneered by DoD research.

Finally, military spending (like government spending generally) is said to stimulate the economy thanks to the multiplier effect.  Money spent by the DoD goes to Soldiers, bureaucrats, contractors, arms manufacturers, researchers, and healthcare providers in the form of wages or payment for service provided.  These people then spend a portion of that added income on other things in the economy – which helps producers of those goods and services expand their business and hire more workers.  Newly hired workers can afford to boost their own consumption spending, further fueling the positive cycle.

It is on this reasoning that modern-day Keynesians like Paul Krugman argue the buildup to WWII was what got us out of the Great Depression.  “The Great Depression ended largely thanks to a guy named Adolf Hitler,” Krugman chuckled in 2012.  “He created a human catastrophe, which also led to a lot of government spending.”  This view is highly controversial in the economic literature, but has nevertheless caught on as a mainstream historical legend.  It’s no surprise that similar arguments were used to augment the case for a 50 percent spike in defense spending in the decade after 9/11.  Just days after the attack, for example, Krugman himself saw a silver lining: “the driving force behind the economic slowdown has been a plunge in business investment. Now, all of a sudden, we need some new office buildings…”  


Each of these arguments are severely misguided.  Whatever their effect on spreadsheet econometrics, bloated military budgets inhibit the economy from producing and distributing things of real-world value that actually improve people’s lives. To see why, we have to understand three basic economic principles.

The first principle is the broken window fallacy.  As Frederic Bastiat first noted nearly 200 years ago, breaking a window causes production to go up: a replacement window must now be created and paid for.  This payment is captured as GDP; but of course, it doesn’t make anyone better off than they were before the window was broken – it merely returns things to the way they were.  Likewise, we could hire federal employees to dig big holes and fill them back up again (something Krugman literally prefers to lower taxes), and employment would technically be created.  But that’s a silly thing to cheer for, because they wouldn’t be employed doing anything worthwhile!  Even if your only goal was to help the unemployed or achieve a multiplier, it’d be better to hand them a simple transfer payment without putting them through the hassle of digging holes.

Furthermore, evidence indicates that the multiplier for defense spending in particular is relatively low, particularly when financed by debt.  An analysis of four studies conducted between 1984 – 2011 put the short-term defense spending multiplier at somewhere between 0.4 to 0.6, and the long term multiplier between 0.6 to 0.9, with no one year having a higher multiplier than 1.2.  The authors estimated that “over five years, each $1 in federal defense-spending cuts will increase private spending by roughly $1.30”.  In other words, defense spending does not create as much valuable second and third-degree spending as is consumed by the initial transaction.

The point is that not all employment or GDP are created equal, and not all spending pays for itself.  It matters what you’re employing people to produce.  Keynesians are right that government spending has a multiplier effect which ultimately increases GDP and employment.  But GDP and employment are only important macroeconomic indicators insofar as they reflect actual improvement to people’s standard of living.  If military production doesn’t do that, it’s not worth buying.

Of course, it is not fair to say defense spending is like digging holes and filling them back up again – it does provide some value.  In fact, as a whole, it’s among the most important things government does.  But this is where our second economic principle comes into play: the law of diminishing returns.

The more you have of something, the less valuable each additional unit of that thing becomes to you.  I might gladly pay $8 for a Five Guys hamburger; if I’m super hungry, I might even buy two at that price.  But if I’ve already eaten two hamburgers in the same sitting, a third one is no longer worth $8 to me. I’ve had enough hamburgers!  Likewise, the larger our nation’s military spending becomes, the less and less additional protection Americans receive for each extra dollar devoted to it.  You can only be so safe!

The value provided from defense spending (or jeopardized by cutting it) must be understood in this context.  It has nothing to do with the importance of defense generally, and everything to do with the importance of marginal increments of defense – which, in turn, depend in part on the strength, number and credibility of the threats the spending defends against.  So let’s compare: “how much defense” does the United States military have, compared to “how much attack” its most powerful potential adversaries have? The U.S. has each of the following:

  • About 6550 nuclear warheads.  Russia has roughly as many; apart from Russia, no other nation has more than 300
  • Over 6,000 M1 Abrams tanks, which are at least as good as the closest Russian (T-90) and Chinese (Type 99) competitors.  Russia has only 800 T-90As, and China has only 500 Type 99s.
  • 7,882 planes and 6,730 helicopters between the Army, Navy, and Air Force; Russia’s in second with 3082 and 973, respectively.
  • 800 military bases in 70 countries around the world.  By comparison, Britain, France and Russia combined have about 30 bases between them.
  • 19 aircraft carriers – about as many as the rest of the world combined.  Japan and France, both allies, are tied for second with four apiece.  Russia and China have one apiece, both far less advanced then the US’s 10 Nimitz class carriers.
  • 54 SSN (attack), 14 SSBN (ballistic) and four SSGN (guided) nuclear submarines, again the most in the world.  Russia has four, five, and four, respectively, while China has five, five, and zero.  
  • An overall budget three times as large as that of its nearest competitor, and greater than the next seven largest combined (of which five are allies).

It bears repeating: you can only be so safe.  The additional defense provided from America’s 5,000th nuke, tank, plane or helicopter is long-since diminished, because such totals simply dwarf those of any prospective threat.  Defense spending as a whole may not be comparable to digging holes and filling them back up again – but at the margins, it’s remarkably similar.

Contrary to stubborn Keynesian mythology, the economic downsides of such wasteful over-provision far outweigh any incidental benefits.  To see how, we need a third principle: the Guns and Butter curve.  In a world of scarcity, producing more of one thing leaves fewer resources to produce other things.  Ergo, each additional dollar spent on defense not only provides less and less added protection from external threats, but also reduces provision of competing priorities by larger and larger amounts.  This tradeoff exists in both the public and private sector.

In the public sector, defense spending contributes to annual fiscal deficits in the hundreds of billions, and a running national debt around $20 trillion.  It reduces funds available to improve crumbling infrastructure, fund collegiate education, redesign a broken healthcare system, or do any of the other things government tries to do.   When the U.S. commits to defending its rich foreign allies as well as itself, it effectively subsidizes those nations’ comparatively robust social safety nets, at the expense of our own dilapidated public services.

Excessive defense spending also necessitates higher taxes, which crowd out private spending on everything else in the economy.  This exacerbates the “broken window” problem by reducing the portion of GDP that actually faces a market test. Private sector spending is inherently more efficient than public, because when people spend their OWN money on things THEY themselves will use, they are motivated to both minimize cost and maximize quality in ways policymakers (spending other people’s money on projects that benefit others) simply are not.  Constant stories of the Pentagon losing or wasting billions on frivolous expenditures underscore this reality, as does the intractable problem of “use-it-or-lose it” budgeting practices throughout the DoD.  The end result is hundreds of billions of dollars-worth of forfeited utility among American consumers, making them worse-off in their own estimation.

These “unseen” trade-offs also challenge the narrative that WWII defense spending ended the Great Depression. So much labor, metal, fuel, construction materials, and other scarce resources were consumed by the war effort that the Americans left behind endured severe economic hardship throughout the war years, including rations on butter, gasoline, clothing, tires, coffee, and other everyday products.  Government posters urged them to “use it up, wear it out, make it do, or do without.”  A command economy with price and rent controls restricted economic freedom but failed to prevent rapid inflation.  Home and auto sales plummeted. Job offerings consolidated into cramped “war production centers,” forcing mass migration to urban areas that uprooted entire communities.  Work hours became longer and riskier – inconveniences often suffered by women and teenagers, who previously had not needed to work at all.  To be sure, all of this boosted employment and GDP (after all, it’s fairly straightforward to achieve full employment with a military draft!)  But because those drafted (or riveting from home) were only employed to defeat faraway threats previously unknown to them, they were merely fixing a broken window.  The actual standard of living did not improve much until after the war – concurrent with a massive reduction in spending, taxes, and wartime economic regulations.

There is also a darker reason excess defense spending hurts the economy.  A core function of war is to blow stuff up; the broken window fallacy is rarely so literal as when bombs and artillery destroy homes and infrastructure.  And as I have argued elsewhere on Realist Review, spending so much on defense may actually impede defense, by enabling unwise foreign entanglements that ultimately invite blowback.  In this way, military spending is not merely wasteful relative to more efficient options, but uniquely destructive to economic growth.  

Another problem is that military spending has diffuse economic costs, and concentrated economic benefits.  War hurts the economy as a whole for all the reasons just described – but it is certainly big business for a select few rent-seekers in the military industrial complex.  When Keynesians argue that military spending creates wealth due to a multiplier effect, they fail to specify that most of the wealth created is gobbled up by such a tiny, well-connected few.

Finally, what of the argument that defense spending pioneers innovation?  In fairness, some of it does; about 12% of the DoD budget goes to Research and Development, and historically this has contributed to some important breakthroughs.  But if the American people want to invest public dollars into technological innovation, it would be more efficient to do so without tethering those innovations to military applications. A primitive internet, for example, may have been pioneered by military funding in the 1960’s – but there it languished for 30 years, connecting only a small handful of computers for a narrow band of applications, until more visionary minds in the private sector were finally allowed to make it useful for the broader public.  And of course, innovation can hardly justify the other 88% of U.S. defense spending, with all its harmful effects.


The United States is far from the only government with an institutional tendency to fund its military too generously.  An extreme example is North Korea, which spends nearly of third of its GDP on defense, despite 70% of its population relying on foreign food aid, and 40% being undernourished.  Another was the Soviet Union, whose perceived need to keep up with U.S. defense spending arguably played an important role in its eventual collapse.  Nevertheless, the U.S. is the only remaining nation that funds a military so exponentially larger than necessary to defend against credible foreign threats.  And although economics alone cannot determine what level of defense spending is appropriate, the case against this DoD boondoggle is sharpened by understanding the economic damage it inflicts.

Beyond what’s needed to deter and repel a far-fetched invasion, defense spending does very little to improve American citizens safety or wellbeing.  The relevant question is not whether the military could find some use for additional spending (it always could), nor whether those uses would make the military marginally better prepared to defend against certain threats (they may).  Rather, the question is whether that much-diminished return in “defense provided” is more important than anything else that $1 trillion might be used for in the economy.  No matter your political persuasion, you should find that proposition dubious.

Andrew is a 2015 graduate of Johns Hopkins University, a former Legal Intern at The Federalist Society, and a future President of the United States. For now, he rants about politics, philosophy, and the Green Bay Packers at






Leave a Reply

%d bloggers like this: