Those who are optimistic about the potential for accommodation between major powers often resort to arguments about the pacifying influence of economic interdependence. If one accepts this view and gives credence to the deepening ties of globalization, it is natural to anticipate that the accommodation of those states assessed as current cases in Part III (China, Russia, India, and Brazil) might proceed more peacefully than the historical cases outlined in Part II. For example, it has been widely argued in American policy and academic circles that the United States and China are less likely to get into serious conflict as China rises because of the extensive trade interdependence between them. As Madeline Albright observed in her announcement of the Clinton administration's decision to renew China's Most Favored Nation trade status, “we have in our era an unprecedented opportunity to integrate the world around basic principles of democracy, open markets, law and a common commitment to peace.”

Proponents of this position tend to argue that that there is meaningful quantitative empirical evidence linking interdependence to peace (or at least to the lessened probability of conflict). Bruce Russett, a primary producer of this quantitative work in support of the liberal position, clearly articulates this stance, drawing the conclusion in a policy-oriented review of his work that conflict is unlikely in interdependent relationships because “if we bomb the cities or factories of a close trading partner – where we also are likely to have heavy private investments – we are bombing our own markets, suppliers, and even the property of our own nationals.”

The trouble with this conclusion is that there is actually very little that most of the quantitative work in this domain can say about the accommodation of rising powers such as China. In the vast majority of cases, the findings that underpin the linkage between interdependence and peace tie measures of trade to the incidence of relatively low-level conflict events. Moreover, to get the statistical power needed to make inferences from these data, the regressions are typically time-series, cross-sectional aggregations of all countries over many years. While appropriate for establishing the general relationship between trade and conflict, almost every aspect of this arrangement undermines the applicability of the findings to the more significant question of major power accommodation in the present age.