Does neorealism offer a convincing account of great power balancing behavior? Many scholars argue that it does not. This conclusion rests on a misunderstanding of neorealist theory and an erroneous reading of the evidence. Properly specified, neorealism holds that great powers place an overriding emphasis on the need for self-help. This means that they rely relentlessly both on arming and on imitating the successful military practices of their peers to ensure their security. At the same time, they rarely resort to alliances and treat them with skepticism. There is abundant historical evidence to support these claims. Since 1816, great powers have routinely achieved an effective balance in military capabilities with their relevant competitors and promptly copied the major military innovations of the period. Case studies show that these outcomes are the product of states' efforts to ensure security against increasingly capable rivals. Meanwhile, the diplomatic record yields almost no examples of firm peacetime balancing coalitions over the past 200 years. When alliances have formed, great powers have generally doubted the reliability of their allies and of their opponents' allies. Thus neorealism provides a solid foundation for explaining great power balancing behavior.

There is broad scholarly consensus that the relative power of the United States is declining and that this decline will have negative consequences for international politics. This pessimism is justified by the belief that great powers have few options to deal with acute relative decline. Retrenchment is seen as a hazardous policy that demoralizes allies and encourages external predation. Faced with shrinking means, great powers are thought to have few options to stave off decline short of preventive war. Contrary to the conventional wisdom, however, retrenchment is not a relatively rare and ineffective policy instrument. A comparison of eighteen cases of acute relative decline since 1870 demonstrates that great powers frequently engage in retrenchment and that retrenchment is often effective. In addition, we find that prevailing explanations overstate the importance of democracies, bureaucracies, and interest groups in inhibiting retrenchment. In fact, the rate of decline can account for both the extent and form of retrenchment, even over short periods. These arguments have important implications for power transition theories and the rise of China.