Someone casually reading the news would have the right to be confused over seemingly contradictory stories on the political preferences of military service members and veterans. According to Gallup, "Veterans Give Romney Big Lead Over Obama," while Reuters claims that "Weary Warriors Favor Obama." Both are true. Understanding why will help dispel some stereotypes about the military while also highlighting that we are in the midst of a generational shift in military attitudes.
The bulk of America's veterans come from older generations, when the draft was in force and military service was the norm and not the exception (at least among white males). Therefore there is a high correlation between being a veteran and being a white male over the age of 65. So most discussions of 'veterans' are necessarily discussions of older white males and it should come as no surprise that this group leans Republican, although that would not make for much of a headline.
Unfortunately in Gallup"s report on the veteran population they only compare their sample of veterans with adult men in general, which provides an insufficient control for age given the differences in the median age of the veteran population and the larger population of non-veteran adult males. In Gallup's own analysis, older age is a significant factor for predicting Republican leanings among whites, and the preferences of older whites in general seem to track the preferences of older vets. However Gallup does not break down its analysis of older whites by gender, or we could see just how much overlap there is between 'older white males' in general, and 'older white males' who are also veterans. Also there are unfortunately not enough young veterans in Gallup's poll to differentiate between the youngest cohort of veterans and those in their 40's. So while the Gallup report is accurate in portraying the attitudes of older veterans, the reported results do not do enough to isolate the effect of veteran status from age and gender nor do they provide insights into the cohort of service members and veterans from the 9/11 era.
Chart courtesy of Jeremy Teigen.
Given that fewer than 1 percent of Americans currently serve in the military and that the proportion of veterans in the population declines precipitously for those under 60 it is understandable that few national surveys capture the opinions and attitudes of younger veterans and active service members. The resulting lack of information on this group has led many to assume that the attitudes of young veterans and service members can be extrapolated from those of older veterans.
However, my study of the active-duty Army, plus analysis of other surveys over the last eight years indicate there are key differences between veteran cohorts and that the opinions of service member may be undergoing a generational shift.
With its American Mosaic poll, IPSOS/Reuters is making a concerted effort to survey this population in the 2012 election cycle. As of July IPSOS/Reuters had reached over 18,000 service members, veterans and their families. Such a large sample of a typically hard to reach population is rare and presents the opportunity to compare younger veterans against older cohorts. There are also enough respondents who report currently serving in the military to further differentiate between veterans and those in active service. In all there are 8580 military and veteran respondents in the IPSOS/Reuters survey whose period of service ranges from pre-1960 to those still in uniform today.
In looking at their attitudes, the results largely match with my previous analysis in that there are some key differences between older veterans and younger veterans and service members. The results of the IPSOS/Reuters data also reveal a few counterintuitive findings that should serve as a caution to those who might view 'veterans' as a monolithic voting bloc.
For starters, many likely assume that for veterans and service members questions of war and foreign conflict or terrorism are the most salient considerations when choosing a candidate. It is also likely that this assumption is what drives candidates and campaigns to seek the endorsement of members of the military. However, when asked about the most pressing issues facing the United States, only a minority of veterans and service members see these as preeminent concerns. Among veterans, only 5 percent see war and foreign conflict or terrorism as the most important problem facing the country. This increases to 11 percent among those actively serving, but is probably still much less than many would assume. Like most Americans, a plurality of veterans and service members view the economy as the country's primary concern.
Similarly, there are few linkages between military service and attitudes on current political debates. When studying the active-duty Army in depth, I found that on most social issues and questions of how the government should spend money, the attitudes of service members largely tracked those of the civilian population. More importantly, members of the Army appear to develop their outlook on social and political issues independent of military service and will often default to partisan cues on many of the political questions of the day. This is an important point to remember when political parties or candidates use veterans for military endorsements: It is likely that the endorser would have had leanings toward the party in question, independent of his or her military service.
... an important point to remember when political parties or candidates use veterans for military endorsements: It is likely that the endorser would have had leanings toward the party in question, independent of his or her military service.
This raises the question of why the military has been seen as reliably Republican for so many years. This appears to be because of a loose cohort effect whereby the political environment during the formative years around the time a person begins military service shapes political party identification in a lasting way, particularly among military officers. Since the Vietnam War this effect has largely tilted members of the military toward identification with the Republican Party. As might be expected a high proportion of the officers who began their career during the Reagan era self-identify as Republican. This proportion decreases slightly among those who joined the military at the beginning of the Clinton administration before going up again at the end of the Clinton presidency and the beginning of the G.W. Bush administration.
However, the effect appears to have shifted for those who entered the military after 2001. As of 2004 there was a surprising level of parity in the party identification of young soldiers and officers. This phenomenon, paired with a significant decrease in support for the Republican Party among service members and veterans between 2004 and 2006, resulted in an unexpected degree of support for Obama during the 2008 election, even though he had not served in the military and was facing a decorated veteran.
It is likely because of this increased parity in the 'veteran vote' that both campaigns are ramping up outreach to the military community, and it is here that the IPSOS/Reuters poll offers some interesting insights into current electoral dynamics. President Obama appears to be garnering support from service members and veterans on issues related to national security, even when the Democratic Party generally still lags behind the Republican Party in perceived effectiveness in this area. When asked which political party has the better plan for conducting the war on terror, most veterans and service members gave the advantage to the Republican Party (an overall gap of 18 percent), with those over 40 expressing the strongest preference for Republican policies. Younger cohorts were more evenly split, with those in their early 30's even showing a slight preference for the Democratic Party in this area.
However, when asked which specific candidate for president, Barack Obama or Mitt Romney, has the better plan for the war on terror the gap among all vets closes considerably to a distance of only 4 percent. Those between the ages of 26 and 40 even showed a preference for Obama over Romney, and 35 percent of those actively serving expressed a preference for Obama's policies compared to 31 percent who think Romney has the better plan. While this may be counterintuitive to the way many perceive the attitudes of veterans and service members, this should not be a surprise given that most Americans have a favorable view of the President's handling of foreign policy.
What this means for the 'military vote' though, may be not much, given the low salience of foreign policy during this election, even among veterans and service members. But the data is interesting in that it shows a level of parity towards the president and his rival that many might not expect, even if it won't be the most relevant factor in how this group votes in the upcoming election. That said, it may be indicative of further movement in the generational shift in military attitudes that appeared to begin almost ten years ago.
Among the youngest service members and veterans in the survey, those ages 18 to 25, there is nearly an even split in party identification with 36 percent identifying themselves as Democrats and 41 percent as Republicans. The advantage is still obviously to the Republican Party, but the gap is significantly closer among the young than it is with older veterans, who average a 30 percent to 48 percent split between Democrats and Republicans, respectively. However, when IPSOS/Reuters asked respondents who they would vote for if the election were held today, the gap in vote preference for the youngest generation of service members and veterans closes even further, with 39 percent and 40 percent choosing Obama and Romney, respectively. Romney's greatest levels of support were found among older veterans, in keeping with Gallup's analysis.
For the most part this is to be expected, as Obama has a significant advantage in the youth vote in the general electorate as well, but to see this preference among young service members and veterans may mean that the era when it seemed that Republicans had a lock on the military vote is coming to a close. It is also worth noting that even when controlling for age, race and gender, whether or not a veteran is currently serving is a significant indicator of positive approval ratings for President Obama. This means that while older generations of veterans remain solidly Republican, new veterans and young service members have started their careers with, and will most likely maintain, a more balanced outlook towards partisan politics.
In the long run this is good news, as it should reduce some of the perceived distance between members of the military and the public they serve. But in the short term it does raise risks to the military's reputation for apolitical service as both parties are likely to increase outreach efforts in an attempt to claim the prestige of the military vote. The active force, constrained by law and institutional custom, can be expected to stay out of the race. However, there may be greater temptation for veterans to enter the fray, claiming to represent the views of the military in the general election. It is therefore essential that Americans have a better understanding of the attitudes of veterans and service members, and to recognize that there is no monolithic 'military vote'.
*Reported results from the IPSOS/Reuters survey are weighted by gender to match the composition of the active-duty force and veteran population.
Jason Dempsey is a career infantry officer in the United States Army and the author of Our Army: Soldiers, Politics, and American Civil-Military Relations. The views expressed here are his own.