Rejoinder to Murdoch and Sandler

In their reply to Oneal and Elrod (1989), Murdoch and Sandler defend the joint product model of alliance behavior and respond to several of our criticisms of their work on NATO. The most important point of this rejoinder is that Elrod and I never questioned the theoretical value of the joint product model, only Forbes, Murdoch, and Sandler's application of it to the North Atlantic alliance. We remain unconvinced that NATO made a significant shift from deterrence to defense in 1967 or 1974. The alliance seems throughout its history to be a uniquely privileged group providing a relatively pure public good. Yet, we agree that burden sharing in NATO has changed in important respects over the years, especially after the late 1960s and early 1970s. Elrod and I argued that two influences were important: the decline of U.S. hegemony and, especially, the pursuit of private benefits by the European allies. At first glance, Forbes, Murdoch, and Sandler's emphasis on impure public goods seems unobjectionable. The growth of the Soviet nuclear arsenal undermined the doctrine of massive retaliation; the U.S. formally urged a shift to a defensive strategy as early as 1962; and the alliance announced flexible response in 1967. But did NATO in fact move from deterrence toward defense? If so, it must be detectable not only indirectly in changing allied defense burdens but also by more direct military measures. The defense of territory requires the ability to stalemate or defeat the enemy, not merely to punish it, but Elrod and I found no evidence that the military balance in Europe tilted in NATO's favor in the decade after 1967. The Europeans, whose security was at stake, did not provide new forces. Their defense burdens declined as the Soviet Union's increased. Consequently, the military expenditures of NATO Europe as a percentage of those of the USSR fell from a peak of 67 percent in 1960