You are using an outdated browser.
Please upgrade your browser
and improve your visit to our site.
Skip Navigation

The Real Costs of the War in Afghanistan

As America's longest war drags on, some Beltway denizens argue that it can be sustained for decades more.

A U.S. soldier after a suicide attack at the Afghanistan-Pakistan border in Nangarhar province in 2014. (Noorullah Shirzada/AFP/Getty Images)

Now in its nineteenth year, the Afghanistan war just won’t end. Negotiations between the U.S. and the Taliban are now dead, according to President Donald Trump. The president, who once clamored for an end to the war, has instead overseen an increase in bombing. Last week, citing rampant corruption in the country, Secretary of State Mike Pompeo cancelled $100 million in aid intended to fund infrastructure projects in Afghanistan, a reminder that the Afghan government stood up by U.S. forces in 2002 still struggles to provide security and stability. The U.S. is staring at the prospect of an even longer presence in Afghanistan—so long that nine former senior U.S. senior envoys to Afghanistan made a case earlier this month that the war actually isn’t that costly and that a significant long-term American troop presence, perhaps even for decades, is feasible.

“U.S. fatalities are tragic, but the number of those killed in combat make up less than 20 percent of the U.S. troops who died in non-combat training incidents last year,” those officials—who include Bush-era appointee John Negroponte and Bush-Obama veteran Ryan Crocker—wrote in their plea not to pull troops out of Afghanistan prematurely. “U.S. direct military expenditures in Afghanistan are approximately three percent of annual U.S. military spending, down by about 90 percent from the high point of the war.” These facts were apparently cause to be sanguine about continued occupation: “The lives and money being expended are serious, but the costs are ones we can sustain for negotiations to result in a sustainable peace.”

There are many ways to measure the costs of something as complicated as war. The ambassadors’ cherry-picking method—to focus solely on deaths of service members and line-item budget costs—is to ignore the considerable indirect costs of war, costs that can’t be wished away. The modern American way of war, with an all-volunteer force and financing raised via debt instead of taxes, has obscured many of the traditionally visible costs of waging war that have often led to popular resistance. Before the American populace—especially the portion asked to serve in this multi-generational war—is involuntarily committed to another 20 years, it’s worth asking what a real, honest cost-benefit analysis of the war might look like.

A baseline annual cost for the United States to continue its war in Afghanistan is approximately ten to 15 U.S. service member fatalities each year. (With 17 fatalities thus far in 2019, this year has been the deadliest year for the U.S. since 2014.) Additionally, the war costs approximately $50 billion per year; the U.S. Department of Defense estimates $45 billion, while others place it at $52 billion.

In the context of other DoD operations and activities—as the ambassadors’ argument places them—these numbers appear low. More U.S. service members did, indeed, die in training accidents than in combat operations. The DoD budget for 2019 approaches $700 billion, and $50 billion might not seem like so much as a share of that. But this annual budget is separate from the Overseas Contingency Operations fund, which has been described as a war-making slush fund and has added $1.8 trillion to military spending since 2001. And clearly neither kitty tells us much about the war’s hidden costs.

The days of war bonds and drafts are far behind us, and with them, a clear and obvious understanding of the costs of war. The Vietnam War essentially ended because of the political costs representatives faced if they authorized more spending for the war effort; when they failed to do so, the White House could no longer sustain the costs, and the complete withdrawal of combat troops began. Today, we have an Authorization for Use of Military Force (AUMF), passed following the 9/11 attacks, which has essentially empowered three successive presidential administrations to expand the war on terror, without much of a check from Congress. That, and the replacement of a universal draft with an all-volunteer force, makes the long-term commitment to the war almost inevitable, with no consequences for politicians to continue to fund it.

As in Vietnam, the cost of the Afghanistan war to veterans—not just in fatalities or visible injuries, but in mental wellness, social adjustment, and economic participation—is significant and growing. Veteran deaths by suicide exceeds service members killed in combat. Even as Afghanistan “winds down,” the suicide rate among younger veterans is increasing substantially. Even service members who don’t see combat at all are still exposed to chemicals from burn pits that lead to long-term disabilities.


Not that the government tracks these costs against the war’s objectives in any meaningful way. Numerous mechanisms have been built to obscure the true monetary costs of our various ongoing wars. Uncle Sam “has never developed a convincing method of reporting on the cost of the wars, and its estimates are a confusing morass of different and conflicting data,” according to longtime security analyst Anthony Cordesman: Simply put, the government “has failed to find any useful way to tie the cost estimates it does release to its level of military and civil activity in each conflict,” much less “found any way to measure the effectiveness of its expenditures or tie them to a credible strategy to achieve some form of victory.”

If we could more fully account for all these indirect costs, would we weigh them? Wars are supposed to be justified, and the sacrifices of the soldiers who serve in them honorable, when (and only when) they are fought for just political objectives that we reasonably believe can be achieved in a short amount of time at acceptable costs. For the U.S., the war in Afghanistan doesn’t meet these criteria.

Everything we know about insurgencies and terrorist groups predicts many more years of violence in Afghanistan. The Taliban are one of the wealthiest armed non-state groups in the world. Weak or developing democracies don’t often defeat insurgencies. Insurgencies last longer in counties with rugged terrain. Insurgencies last long if they have foreign backers or can find relative sanctuary in neighboring countries. Counterinsurgencies are more difficult in ethnically heterogenous societies. Poor socioeconomic development and little past experience with democracy also undermines external democratization efforts.

Preexisting conditions in Afghanistan make a successful democratization effort there nearly impossible, and policy decisions have made a successful outcome even more difficult. The risk of losing political progress with a withdrawal needs to be weighed against the fact that government and coalition forces are now killing more Afghan civilians than the Taliban are. The sacrifice of U.S. service members was unlikely to lead to a successful democratization effort when there were more than 150,000 soldiers from NATO-member countries and we were spending hundreds of billions of dollars. It is even less likely now, with no clear strategy for success.

“A short jump is certainly easier than a long one, but no one wanting to get across a wide ditch would begin by jumping half-way,” wrote Carl von Clausewitz, the oft-quoted nineteenth-century military strategist. Meaning, military force should be wielded decisively and overwhelmingly to achieve a stated objective and then stop once it is achieved. Advocates of staying in Afghanistan indefinitely are, on this view, attempting to cross the Grand Canyon by jumping toward the middle and seeing what happens. Soldiers are being sacrificed for an objective that is unlikely to be achieved. That the real costs of the war can’t be fully tallied is less important: Under these conditions, even one more casualty can’t be justified.