Windypundit

A couple of weeks ago, I posted about the Unreal Liberal Argument for Conscription, which is basically that the U.S. would be less likely to engage in military adventurism if more people had skin in the game because their children could be drafted. My response was essentially that without the ability to force young men to be soldiers, the government would have to be far more careful about how it picked and fought its wars, an argument I backed up with some rough numbers indicating that we lose far fewer soldiers to war since conscription ended in 1973.

My tweet announcing this post was re-tweeted by Radley Balko (who is probably the most famous person who follows me) which lead to a few interesting responses:

Kevin Wilson tweeted:

There’s also the moral argument that conscription is basically slavery: http://econlog.econlib.org/archives/2011/05/how_could_the_d.html …

XM in Dallas tweeted along the same idea:

Great piece. But am I being obtuse by suggesting that we should keep saying out loud that slavery is wrong?

I think that conscription is slavery too, and I’ve said so before. I didn’t mention it explicitly this time because I didn’t think it was a useful response to David Sirota’s argument for conscription. Sirota’s presumably not an evil person, and if he were amenable to the idea that conscription is slavery, he probably would not have made the argument that he did. The people I wanted to reach, those who are swayed by his argument, probably don’t think “it’s slavery” is a compelling point.

But why don’t they? Why does the link between slavery and conscription — which seems practically self-evident to me and others — not seem sensible to so many other people? Bryan Caplan’s article that Kevin Wilson provided in his tweet offers one plausible explanation:

It’s tempting to dismiss all this as doublethink, but after many years of reflection I think I finally figured out what most people are thinking.  Namely: They implicitly regard slavery not as mere involuntary servitude, but as low-status involuntary servitude.  Since most of us honor, respect, and even adore all our soldiers, conscripts have high status – and therefore can’t be slaves.  From this point of view, saying “conscription is slavery” isn’t righteously standing up for the rights of conscripts; it’s wickedly denying them their high status.

That strikes me as quite likely to explain some people’s reactions, perhaps especially on the right. It explains how people can express admiration for our soldiers while simultaneously arguing that we should take away their freedom.

When it comes to liberal advocates of conscription, however, I think that there’s a different dynamic at work. Consider part of an argument made by my (somewhat less libertarian) co-blogger Ken in the comments:

Do we really choose our wars more carefully now? It seems we rushed headlong into the Iraq war with almost no legitimate justification. The monetary cost certainly didn’t seem to have had a detrimental impact on the decision to invade. Those costs weren’t even budgeted for at the time.

Even if we don’t explicitly debate the benefit-cost calculations of wars explicitly, we have had extensive debates about the size and structure of our military forces. Our military may be larger than most of the rest of the world combined, but in terms of total soldiers under arms, it’s smaller than it used to be. As the cold war ended, we took the opportunity to stand down about 1/3 of our active duty personnel, and there was a lot of discussion of how this would limit our strategic options. The invasion of Iraq strained the limits of our ability to fight battles abroad. At the risk of invoking Donald Rumsfeld, we can only warmonger with the army we have.

A completely different criticism comes from Jon S, who tweeted,

Who knew that only American deaths count??

Every death counts, of course. But I was responding to the argument that re-instituting conscription would make Americans more likely to oppose rampant militarism because they or their loved ones could be drafted into a war. The death rates of our foreign opponents (or even our allies) wouldn’t figure into that argument, so I didn’t take them into account in mine.

Jon S may have been trolling me, because a little later he tweeted:

Witnessing/participating in horrific violence is cost of war,eg PTSD. If cost spread over whole pop, less likely 2 support it

It’s hard to see how conscription in the U.S. could spread around the cost incurred by people from other countries, so he’s clearly talking only about the affect of witnessing/participating in horrific violence as incurred by U.S. participants, which is what I was doing too, and for the same reason.

(It’s possible I misconstrued his tweet and he was really objecting to my only counting fatalities and excluding other harm. I used war deaths as a proxy for total social cost of historic and current wars because I don’t know where to find reliable estimates for those costs.)

I did give the issue of foreign cost some thought as I was writing my post, especially the part about how giving potential soldiers the option to refuse to serve was forcing the military to spend more money on equipment and training. That effect protects American lives, but foreign solders wouldn’t benefit from it. In fact, the enormous investment in American soldiers makes them more effective at killing the enemy, which could conceivably offset the beneficial effect of saving American lives and fighting fewer and smaller wars. If the volunteer military has saved 90,000 American lives but American soldiers have killed an extra, say, half-million of our enemy, then ending conscription has arguably increased the amount of misery in the world.

For purposes of my last post, however, I decided to count the cost of war only in American deaths because (1) as I said, the deaths of foreigners were not part of the argument I was countering, and (2) it’s a much more complicated question requiring a lot more work.

Nevertheless, it is a fair question, so let me give it a shot. Regarding the first problem, let’s use the basic rules of welfare calculations and assume that everybody counts, and everybody counts equally. Some people may very well think that the lives of American soldiers are worth more than the lives of a bunch of terrorists hiding in caves, but that would be rigging the analysis: If you assume American lives are worth more than anybody else’s, then why bother to do the analysis at all?

The second problem is trickier to solve. For one thing, we don’t have good estimates of how many of the enemy were killed in some of our wars. The estimates of the number of North Korean combat deaths apparently spans a range of many hundreds of thousands, as do the estimates of North Vietnamese dead. The numbers for Iraq and Afghanistan are apparently even murkier, especially since people with political interests have interests in understating or overstating the estimate.

Another problem is that I don’t know how many enemy deaths to attribute to American combat action. Lots of North Koreans were killed by South Koreans. And lots of Iraqis were killed by regional militants rather than American soldiers, so then you have to ask how many of those killings would have taken place if we hadn’t invaded. Some of these answers are knowable, but it would take a lot more historical research than I can do.

Yet another problem is that some of these wars arguably had a net benefit. The UN apparently estimates that Operation Restore Hope in Somalia helped support humanitarian intervention that saved 100,000 lives. After North Vietnamese forces conquered South Vietnam, hundreds of thousands of people were executed, died in concentration camps, or died trying to flee, and the Khmer Rouge killed over a million people in Cambodia. So how many lives did we save by not allowing North Korea to do the same thing to South Korea? How many lives, if any, did we save in Bosnia-Herzegovina or Yugoslavia by preventing some unknowable amount of ethnic cleansing? I have no clue how to account for lives saved by warfare, so I won’t. I’m only going to count the cost.

For the time period I was considering (World War II to the present) it’s not unreasonable to make the simplifying assumption that all our wars were wars of choice. We could have kept our troops at home instead of sending them to places like Korea and Vietnam and El Salvador and Afghanistan and Iraq. That means we can reasonably attribute all American combat deaths from those wars to decisions made as part of our national security policy. The same cannot be said of the deaths of our enemies, who may have been killed as a result of non-American action.

To avoid turning this into a life-long research project, I eventually settled on a simple methodology. I would visit Wikipedia’s page on United States military casualties of war and click through to the Wikipedia page for each war after World War II to look up the number of enemy casualties, choosing low or high estimates conservatively with respect to my hypothesis. I’ll fill in a few other numbers from other sources. (If you have better data handy, let me know.)

Counting only Korea and Vietnam, I get a low-bound estimate of 817,000 enemy dead before the United States ended conscription. (The high estimate is 1,915,000 dead, not including 1.5 million civilian deaths on both sides of the Korean conflict.)

In the post-conscription era, I’ll stick to the high-end estimates and include 20000 dead in the El Savadoran Civil war, 70 in our invasion of Grenada, 60 in the bombing of Libya, 3500 in our invasion of Panama, 29,600 in the first Gulf War, 1000 in Somalia (from Operation Irene, the Blackhawk Down incident), 25,000 in Bosnia-Herzgovina, 11,000 in Yugoslavia, and all 13,000 guerrilla soldiers killed in the Columbian conflict so far. I’ll assume all deaths were the result, directly or indirectly, of American involvement.

Wikipedia says there are no generally accepted numbers for the number of casualties from the War on Terror in Afghanistan and Iraq. For Afghanistan, I’ll us the estimate by Jonathon Steele of the Guardian (no shill for United States militarism) of 50,000 dead. Finally, for Iraq, I’ll use the death toll of 655,000 from the Lancet‘s survey of Iraq War casualties, which includes a large number of people killed by militant attacks against the population. (There was a higher estimate by another organization, but the Lancet‘s work was peer-reviewed.)

That brings the grand total to 818,000 non-U.S. dead, which is coincidentally only a thousand more than than during the conscription era. Since the post conscription period was about 1/3 longer (29 vs. 38 years) it ends up being about 25% less deadly for people outside the U.S.

This is not, of course, a very robust finding, and neither was my previous analysis of American war deaths, even though it found a much higher ratio. I have no idea how to determine significance. I’m pulling numbers from dubious sources and disregarding conflicting estimates. I’m using a proxy measurement for the total social cost, and I’m not controlling for other independent variables — most obviously, the end of the Cold War.

(A better approach to studying the subject would probably not rely on statistics. Instead, we should examine the actual historic decision making behind all conflicts entered into — and all conflicts avoided — during the period before and after the end of conscription. This is well beyond my capabilities, but I suppose historians have studies the process at various times and places.)

That said, I do have a plausible theory — that forcing the government to pay the full market price for military service (rather than allowing it to confiscate the labor, freedom, and lives of American soldiers) forces it to protect soldiers more thoroughly and use them more wisely. The numbers we have available seem to support my theory, and they don’t support the theory that eliminating the draft has caused an increase in militarism.

There’s also the question of whether conscription would really “work” by spreading the cost around. One of the criticisms of conscription during the Vietnam war was the ease with which non-disadvantaged people could get deferments or find other ways to avoid being drafted. What reason is there to believe it would be different this time? Unless you really think progressive forces would have the political power to establish a truly class-neutral conscription system. In which case, why not use that substantial political power to fight militarism directly?

Finally — and the only thing that matters to some of us — an all-volunteer army avoids the moral blight of slavery. Seriously, when has slavery ever been the solution to a progressive social problem? Should we encourage higher wages by randomly forcing hundreds of thousands of Americans to work in factory assembly lines or serve in battalions of migrant farm workers? Could we end the sexual exploitation of women by forcing everyone to sell sex? Maybe we could get better healthcare by forcing a larger percentage of the population to become doctors and nurses? And who would get to make those decisions?

I think the all-volunteer system is an important way to ensure that those who serve in our county’s military have a strong say in their conditions and terms of service. And I think that in turn will discourage the use of our military forces for dubious adventurism. Which is, I think, something we should all want.

I’ll always have a warm spot in my heart for the liberal/progressive mindset because I grew up in the immediate aftermath of the 1960′s. The era brought lots of changes for the better — the free speech movement, the expansion of civil rights, Miranda and Gideon — but the one that most vividly affected me at the time was the abolition of the draft. I was only 8 years old — too young to have any understanding of war or dying — but just the idea that I could be forced to leave my friends and family to spend years in the army was frightening enough.

Conscription ended in 1973, but the government required young men to continue registering for the draft until the selective service system was shut down in 1976. That period of respect for freedom lasted all of four years, until the Soviets invaded Afghanistan in 1980, and President Carter re-instituted selective service registration to show the Soviets that we were really seriously angry at them. (The fact that the Soviets have since been driven out of Afghanistan and we invaded it ourselves a few years ago would seem to imply that selective service registration is no longer necessary, but nobody listens to me.)

Since then, the United States has become the single preeminent superpower in the world. No other nation comes close. We basically won the war for global domination. And yet there are still those calling for a return to conscription.

Surprisingly to me, these calls come not from war-mongers on the political right, but from the ideological descendants of the same political liberals who so effectively opposed conscription in the ’60′s and ’70′s. I just noticed an example of this from David Sirota at In These Times in an article titled “The Military’s 40-Year Experiment”:

In operations across the globe, the all-volunteer military has been employed by policymakers to birth what Gen. George Casey recently called the “era of persistent conflict.” Four decades later, we therefore have an obligation to ask: How much of the public’s complicity in that epochal shift is a result of the end of the draft?

This is the beginning of a common bring-back-the-draft argument: That the reason we live in a time of seemingly continuous war is that not enough Americans care enough to oppose it, and they don’t oppose it because they know that it would mostly be fought by other people’s children. Bringing back the draft would mean no one was safe from the consequences of war, which would make us think more carefully as a nation before going to war.

(The idea of making things worse to raise public opposition to make them better is not unusual among those seeking social change. The movie Amistad portrays some abolitionists as wanting the Africans to lose in court because it would help build support for abolition, terrorists often hope to incite reprisals from their targets that will serve to send more people into their arms, and both white and black racial separatists in America have tried to foment racial war, each thinking it would rally their race to their side.)

Sirota supports his argument by looking back to the Nixon administration’s thinking about how to end the draft:

[A] look back at some lost history shows that today’s public acquiescence to militarism was exactly what the government wanted when it ended the draft.

That loaded term—“militarism”—was, in fact, a prominent part of the 1970 report by President Nixon’s Commission on an All-Volunteer Force. In its findings, the panel worried about “a cycle of anti-militarism” in a nation then questioning America’s increasingly martial posture.

Noting that “the draft is a major source of antagonism” toward the growing military-industrial complex, the report praised the fact that “an all-volunteer force offers an obvious opportunity to curb the growth of anti-militaristic sentiment.”

Nixon’s commission did devote some empty rhetoric to downplaying “the fear of increased military aggressiveness or reduced civilian concern” about military actions in the event of an all-volunteer force. But the report’s political conclusions were clear: By disconnecting most Americans from the blood-and-guts consequences of war, the end of the draft would “decrease dissent stemming from conscription” and “close one of the channels” of anti-war organizing.

It’s fascinating that people in the Nixon administration thought this, but both theory and history demonstrate that they (and David Sirota) were missing a big part of the picture. Sirota is also proceeding from a false assumption about our history of warfare.

Today, such conclusions read like prophecy. Though polls showed that many Americans opposed the Iraq War, that invasion and occupation was historically unprecedented in length and yet never generated the kind of mass protest that earlier shorter wars evoked.

So Sirota’s evidence of how bloodthirsty we are is that the Iraq War was “unprecedented in length”? This is disingenuous. As I write this, the Iraq war is only a little longer than the Vietnam War (and maybe shorter, depending on when you want to say Vietnam started) and the U.S. death toll is just under 4500. By comparison, the U.S. death toll in Vietnam was more than that in every year from 1966 to 1970. In 1968 alone, almost 17000 American soldiers lost their lives.

Same thing for the Afghanistan War. Same thing for all the forward deployments to far-flung bases and one-off missions.

David Sirota is writing as if we’re living through a time that suffers unprecedented levels of war, but he’s mistaken in that premise. The last dozen years are more violent than the dozen years that preceded them, but we’re still living in a time of relative peace, and the wars we’re fighting are producing relatively few American casualties.

Let’s consider the numbers: The gigantic death toll of World War II would skew the numbers in a way that’s hard to think about, so for the sake of argument let’s set aside those 400,000 dead soldiers and consider the period from the end of World War II to the end of conscription in 1973. In that 29-year period, we had the Korean War, the Vietnam War, and a number of small deployments that lead to the combat deaths of almost 95,000 American soldiers, for an average of a little over 3200 deaths per year.

After conscription ended in 1973, no American solders died in war for the next 6 years. Things got only a little more violent over the next two decades, with about 400 more combat deaths in the ’80′s (mostly Beirut) and the ’90′s (mostly the first Gulf War). Even with the explosion of violence from the War on Terror (6700 U.S. soldiers dead and still counting) the average annual combat death rate for the post-conscription era is only about 200 per year, or 1/16th the rate during conscription.

Even if we accept Sirota’s implied hypothesis that the War on Terror since the 2003 invasion of Iraq represents the new normal, that’s an average death rate of about 670 per year, or about 20% of the rate during conscription. In the worst year of the War on Terror, 2007, we had 1021 combat deaths, still less than one third of the average during conscription. (By comparison, drivers on Illinois highways have become familiar with electronic signs proclaiming that 957 people died in traffic accidents in this state in 2012.)

Yet Sirota is clearly right about Americans being less engaged with the war:

The pattern suggests that in the absence of conscription, dissent—if it exists at all—becomes a low-grade affair (an email, a petition, etc.) but not the kind of serious movement required to compel military policy changes. Why? Because as former Defense Secretary Robert Gates put it, without a draft “wars remain an abstraction—a distant and unpleasant series of news items that does not affect (most people) personally.”

And yet despite this apparent apathy, the statistics show that war casualties are far lower in the post-conscription era. That doesn’t keep Sirota from reaching an idiotic conclusion:

Well-meaning people can certainly disagree about whether a modern-day draft is a good idea or not (and it may not be). But forty years into the all-volunteer experiment, it is clear that ending conscription was as much about giving citizens the liberty to abstain from as about quashing popular opposition to martial decisions. By design, it weakened our democratic connection to the armed forces—a connection that is the only proven safeguard against unbridled militarism.

Ending conscription may have “weakened our democratic connection to the armed forces” as Sirota believes, but judging by the stunning decline in the death toll, there must be something that caused the U.S. to become far less militant.

An economist wouldn’t have a hard time identifying a likely cause. (And conservative economist Milton Friedman was one of the key figures arguing for an all-volunteer military force.) Fewer Americans may have a personal connection to the costs of warfare now that they and their friends and family can’t be drafted, but ending conscription vastly increased the influence of those Americans who care the most about the human cost of our wars: The soldiers themselves.

In market terms, under conscription the price of a soldier’s labor was held artificially low because draftees had no right to refuse the deal and therefore they had no bargaining power. But since the end of conscription, no solder has joined the military against their will. Every single soldier must consent to the risks of combat. And that consent now comes at a much higher price.

A big part of that price has been paid not in salary and benefits but in a reduction in the rate at which soldiers are sacrificed to combat. Just as corporations facing high labor costs will switch to more capital-intensive operations, the increased bargaining power of American solders has forced the U.S. military to invest heavily in equipment and training to make soldiers more effective with less risk.

An all volunteer army isn’t going to put up with shoddy equipment and poor support. They want reliable guns, working radios, GPS guidance, body armor that can stop an enemy bullet, vehicles that can take a rocket hit or an IED, good battlefield intelligence, air support, decent food, medical care, and internet service. There’s a reason it costs over $800,000 per year to put a soldier in Afghanistan, and it isn’t the soldier’s pay.

Note that all this doesn’t mean that the end of conscription increased the cost of warfare. The difference is that under conscription the cost of warfare was unfairly borne by the soldiers, whose labor, safety, and lives were confiscated for the public good. Now that those soldiers’ services have to be purchased at market rates, the cost is more properly borne by the American people on whose behalf they are fighting.

I suppose you could argue that it’s unfair to compare the present low-intensity wars against the major land conflicts in Korea and Vietnam, but I think that just proves my point: The folks who choose our wars are a lot more careful when when they have to pay the full market price for warfighters.

I’m not arguing that the War on Terror is insignificant, but it is a huge improvement on the horrors of the past. And keep in mind that having wars which are “an abstraction—a distant and unpleasant series of news items that does not affect (most people) personally” is a good thing. The alternative is a much higher human cost.