Does it matter that so few congressmen are veterans?

Former senators Frank Lautenberg and Bob Dole, right, tour the National World War II Memorial in Washington. (Lauren Burke/AP)
Frank Lautenberg, who passed away this summer, was the last of 115 World War II veterans who served in the U.S. Senate. To the best of my knowledge, there will be only 12 U.S. senators who have experienced active military service in the 114th Congress. Only one in five members of the current House of Representatives were active-duty military. By contrast, during most of the Cold War, 70 percent of the U.S. Congress were veterans, with the peak coming in 1977 (80 percent).

Does this matter for policy making? There is some research suggesting that it does, most notably the work by Peter Feaver and Chris Gelpi. Feaver and Gelpi establish the following regularities (see especially this book and this chapter-length update):
— On issues that concern the use of force and the acceptance of casualties, the opinions of veterans track more closely with those of active military officers than with civilians.
— The U.S. initiates fewer military disputes when there are more veterans in the U.S. political elite (the cabinet and the Congress).
— The U.S. uses more force in the disputes it initiates when there are more veterans in the U.S. political elite.
— Veterans are less likely to accept U.S. casualties for interventionist uses of force than for “realpolitik” uses of force.
There are other studies that find that the opinions of veterans are more in tune with realpolitik logic than are the opinions of those without military experience. For example, Geoffrey Wallace finds in a survey experiment that veterans are significantly more likely to support torture compared to civilians without any prior military background and that, unlike civilians, veterans are not moved by the information that torture violates international law.
On the other hand, William Bianco finds (gated) that veteran status exerts only a minor influence on the voting behavior of congressmen. He concludes that the shift away from congress members with military experience is unlikely to influence legislative outcomes much. This study does not look at cabinet members and does not consider how veterans may set the legislative agenda differently. Still, it is important to keep in mind that the differences in attitudes between veterans and non-veterans, while consistent and significant, are not very large in any of the studies that I have seen. Nevertheless, even small differences can be quite important when considering outcomes that are as significant as decisions to use force.