Who Serves in the Military Today?

Three of the four candidates in the upcoming election have a son who has either served in Iraq or soon will: Jimmy McCain, Beau Biden, and Track Palin. (And the children of the fourth candidate, Barack Obama, are a bit too young for military duty.)

Is this sheer happenstance?

I am guessing that when Obama was preparing to pick his running mate, it was important to counter John McCain‘s military bona fides — and Joe Biden fit the bill at least in some small part because his son Beau is a captain in the Delaware Army National Guard, soon to be deployed to Iraq. When McCain chose his vice-presidential candidate, Sarah Palin‘s chances certainly weren’t hurt by having a son who’s an Army Pfc. about to be sent to Iraq.

If you randomly take any four American families, it would certainly be anomalous if three of them had a son in Iraq. (The U.S. military currently has about two million people in uniform.) But isn’t it even more anomalous that three of four families like these — i.e., families of considerable means — have sons in Iraq? Isn’t the modern military full of men and women from low-income backgrounds, with a far higher minority representation than in the general population, who join up only because they have no other viable career possibilities?

That is certainly a piece of conventional wisdom that I have heard voiced; which is why a new report titled “Who Serves in the U.S. Military? The Demographics of Enlisted Troops and Officers” is so surprising. It was compiled by Shanea J. Watkins and James Sherk at the Heritage Foundation’s Center for Data Analysis. I suspect that the Heritage Foundation’s imprimatur will raise skepticism among some readers, and I have several qualms myself with what is said and not said in the report, but the facts are very compelling.

The report measures the demographics of military personnel against the general U.S. population in four areas: household income, education level, racial and ethnic background, and regional origin. Here is the most surprising picture in the report:


So 50 percent of the enlisted recruits (i.e., not including the officers’ corps) come from families in the top 40 percent of the income distribution, while only 10 percent come from the bottom 20 percent. It is worth noting that the income information here is not perfect: the data do not include actual family income for each recruit, but rather use the median household income of the recruit’s home census tract. But still, one look at that graph tells you that the conventional image of a military full of poor kids doesn’t reflect the reality.

“These trends are even more pronounced in the Army Reserve Officers’ Training Corps (R.O.T.C.) program,” reads the report, “in which 40 percent of enrollees come from the wealthiest neighborhoods — a number that has increased substantially over the past four years” (i.e., since the September 11 attacks).

Here are some of the report’s other claims:

1. “American soldiers are more educated than their peers. A little more than 1 percent of enlisted personnel lack a high-school degree, compared to 21 percent of men 18 to 24 years old [in the general population].”

2. “Contrary to conventional wisdom, minorities are not overrepresented in the military service.”

3. “The facts do not support the belief that many American soldiers volunteer because society offers them few opportunities. The average enlisted person or officer could have had lucrative career opportunities in the private sector.”

Point No. 1, while technically true, is also misleading. As the report states elsewhere, “The military requires at least 90 percent of enlisted recruits to have high-school diplomas” (not counting GED’s) and, furthermore, the Army itself requires a high-school diploma or equivalent, with a 2.5 G.P.A.

So high-school dropouts are, for the most part, not getting into the military. In fact, if you consider “low education” a proxy for “low income,” that would seem to explain most of the high-income effect we see in the graph above. This doesn’t make the graph any less true; it just makes the report’s language needlessly boastful.

Point No. 2 is particularly interesting, especially as you dig further into the report’s data. Whites and blacks make up almost exactly the same percent of the enlisted personnel as they do in the general population.

The recruit-to-population ratio for whites is 1.06, and for blacks it is 1.08. Hispanics, meanwhile, are significantly underrepresented among enlisted personnel, with a recruit-to-population ratio of just 0.65. (It should also be said that this entire report groups together personnel from all four service branches, which means that the aggregate numbers do not necessarily represent any one of the branches separately.)

It’s also interesting to note that blacks are overrepresented in R.O.T.C. commissions, with a 1.21 officer-to-population ratio, compared to 1.02 for whites. United States Military Academy graduates, however, are a different story entirely. Just over 80 percent of West Point graduates are white (a 1.12 officer-to-population ratio), while only 5.5 percent are black (a 0.5 ratio). Also, nearly 18 percent of West Point cadets come from a family with a household income of more than $100,000. Granted, West Point is an elite institution and is bound to attract elites.

There’s a further important point that can’t be found in this report but can be found in another one, which compiles race-specific U.S. military fatalities in Iraq and Afghanistan. As of March 1, 2008, there were 2,964 white fatalities in Iraq, representing 74.8 percent of the total; in the general population, meanwhile, whites in that age cohort make up about 62 percent of the population, so whites are overrepresented among Iraqi fatalities. Blacks and Hispanics, meanwhile, are both underrepresented; the same is true in Afghanistan.

Point No. 3 is almost an ideological argument rather than a factual one. But still, this much is clear: when discussing the U.S. military in the aggregate, the common notion that the military is a stop of last resort, increasingly staffed by low-income desperadoes with slim future prospects, cannot be right.

If the report has one significant ideological point to make, it’s that military participation has a huge patriotic/service component that is commonly overlooked, especially in portions of the country where military representation is far below average. (In the Northeast, for instance, the recruit-to-population ratio is just 0.73, compared to 1.19 in the South.)

We obviously haven’t heard the last word on patriotism or service in the current campaign. And many of the words to come will certainly be loaded. If nothing else, here’s hoping that people — no matter which side they’re arguing — will take a look at some of the numbers in this report before leaping to conclusions.

[Note: I recently discussed this topic on The Takeaway.]

Leave A Comment

Comments are moderated and generally will be posted if they are on-topic and not abusive.



View All Comments »
  1. JakeR says:

    Tom Best is concerned that Obama hasn’t served (presumably in wars for which he was eligible) but who was a little long in the tooth (31) when Gulf War I started. There’s no draft now, as there was when George W. Bush, Dick Cheney, et al. in the current administration and among the neocons and other chicken hawks such as Rush Limbaugh who are approximately my age and who managed to avoid the Vietnam war in which I served as a volunteer. If Mr. Best has not served in time of war, we can discount his quibble entirely. If he has, I salute him but nonetheless hold Obama’s age is a reasonable excuse.

    Adam S raises questions that deserve answers. As my grampa used to say, “Figures don’t lie, but liars figure.”

    Well-loved. Like or Dislike: Thumb up 5 Thumb down 0
  2. Tristan says:

    Tom (1), do you really want you Harvard graduates enlisting in the military?

    Thumb up 2 Thumb down 0
  3. Chance says:

    I don’t have a study or numbers to back up my opinion, but my experience in 8 years of service was that the vast majority of my fellow enlisted were enlisting for college money (or to pay off college loans) because they could not easily afford it otherwise. This motivation was followed by being a young mother or father. I am not suggesting patriotism was not a factor, but for most it was not the primary factor.

    As far as whites being over represented in fatalities, I have always heard that this is mainly a holdover from earlier decades where (rightly or wrongly) it was believed that minority soldiers were being used as cannon fodder. When the all volunteer military was instituted, many minority enlistees chose to serve in service support units instead of combat units.

    Well-loved. Like or Dislike: Thumb up 5 Thumb down 0
  4. AS says:

    I would guess that the overall proportion of Hispanic representation is rather high for the population in the Army, based on my personal experience–and that the Air Force and Navy–which are rather white in comparison–bring down that statistic considerably.

    Another thing this doesn’t address is whether they’re counting as “some college” online courses that soldiers are able to take to up their promotion points. This is a great thing, but doesn’t really prove their point that soldiers are already in a good social position before they enlist. While there are people who join for purely patriotic reasons, I would say over half of the Army enlisted people I know joined to get out of a crappy situation economically or socially.

    Well-loved. Like or Dislike: Thumb up 5 Thumb down 0
  5. Tom K. says:

    http://en.wikipedia.org/wiki/Household_income_in_the_United_States The bottom fifth have fewer children than the middle and top. It won’t account for everything but it helps adjust for the skewed chart, I doubt it’s as drastic as it seems. It would have been better to see a 6th quintile $125-250k

    Thumb up 3 Thumb down 1
  6. Ezzie says:

    When this came out I noticed something particularly interesting about the maps that I think is worthy of note. Money quote:

    What is striking is that on the graphic which shows the state numbers of enlisted personnel the bottom rungs are filled almost entirely by liberal states (color is from the 2004 election):

    50) [red] North Dakota (48th ROTC, 41st Academy)

    49) [red] Utah*

    48) [blue] Rhode Island

    47) [blue] Massachussets (43rd ROTC, 44th Academy)

    46) [blue] New Jersey (40th ROTC)

    45) [blue] Connecticut

    44) [blue] New York (47th ROTC)

    43) [blue] Delaware

    42) [blue] Minnesota (41st ROTC, 40th Academy)

    41) [blue] Vermont

    40) [blue] California (50th ROTC, 43rd Academy)

    * Presumably, Utah has low numbers because the Mormon population tends not to join the military.

    Of 19 blue states in the 2004 election, 9 are in the bottom 11. Of 31 red states, just two are. (Actually, after [red] Mississippi at 39 come [blue] Illinois, Pennsylvania, and Maryland – making it 12 of the bottom 15.)

    Thumb up 2 Thumb down 0
  7. holycalamity says:

    To follow up on Adam S.’s comment, I’d like to see if/how the numbers differ for recruits since 2003. It used to be that military service was a way for middle class kids to earn money for college while seeing the world and getting some training, and the likelihood of actual combat was low. It wouldn’t surprise me to see the numbers change since combat became a very real possibility, and the services are desperate to meet recruitment numbers and racing to the bottom.

    And I agree with MikeM at #7 – they seem to be playing games with the vague median income of “neighborhoods” – the same census tract could contain both wealthy neighborhoods and the proverbial “wrong side of the tracks”. My purely anecdotal observation from growing up in the South is that in rural areas, wealthy neighborhoods often abut poor ones, skewing income data even in small areas.

    Anyhow, I’ve learned to take any study published by right wing think tanks with a sack of salt. They’re just like real academics, but they think peer reviews are unpatriotic and elitist.

    Thumb up 4 Thumb down 2
  8. holycalamity says:


    I find that red state/blue state designations tend to be misleading. If you look at the 2004 election maps created by the University of Michigan (http://www-personal.umich.edu/~mejn/election/) you’ll notice that voting patterns aren’t defined so much by region, but by population density.

    Blue States overall have a higher concentration of population in cities, while Red States have larger rural populations – basically, it’s less a matter of coasts vs heartland, but city/suburbs vs rural.

    I think the takeaway is that the military draws more from rural areas than cities.

    Thumb up 3 Thumb down 0