Correspondence: Decline and Retrenchment—Peril or Promise?

To the Editors (Kyle Haynes writes):

Paul MacDonald and Joseph Parent's article "Graceful Decline?" offers a clear, parsimonious theory of great power retrenchment that helps fill a massive gap in international relations scholarship. ¹ Through comparative case studies and "coarse grained" statistical analysis, MacDonald and Parent argue that the degree of a state's decline often explains the form and extent of its retrenchment. They then show that retrenchment is a surprisingly common and effective response to relative decline. MacDonald and Parent correctly point out the myopia of the "pessimistic" structuralist dogma that simply dismisses retrenchment as an impractical and dangerous strategy that only accelerates decline by signaling weakness and creating additional vulnerability (pp. 13-18). ² Their spare neorealist model goes a long way toward repairing this deficiency. As a first cut, it improves on the existing literature while facilitating progressive future research on the topic. Still, a number of theoretical and conceptual problems undermine their argument and compromise their results. Below I discuss three issues with MacDonald and Parent's theory of retrenchment.

Power Transitions and Policymakers' Responsiveness

MacDonald and Parent operationalize decline as an ordinal power transition in which a rising state overtakes a declining state in terms of gross domestic product (GDP) and retains this new lead for at least five consecutive years. They justify this focus by claiming that such transitions are particularly "dramatic" events that should force policymakers in the declining state into a thorough strategic reassessment, potentially leading to retrenchment (p. 21). These are "most likely" cases for their neorealist model, as less dramatic periods of decline may fail to generate policy shifts with similar promptness.

This focus on ordinal power transitions is unjustified and may systematically ignore important periods during which retrenchment dynamics should be operative. For MacDonald and Parent, overextension and decline produce fiscal insolvency, which in turn induces retrenchment (p. 19). ³ But importantly, the years following an ordinal power transition do not generate insolvency more rapidly or with greater severity than other periods in a state's decline. An ordinal power transition thus does not induce retrenchment because it engenders uniquely potent material incentives for withdrawal. Rather, MacDonald and Parent seem to imply that policymakers will belatedly recognize exogenous and preexisting incentives for retrenchment only in the wake of such psychologically jarring transitions.

Although an ordinal power transition may indeed be a dramatic event in hindsight, MacDonald and Parent's empirical focus requires a demonstration that relevant policymakers considered it as such. Problematically, policymakers are more likely to respond to exogenous shocks such as economic crises, unexpected foreign policy setbacks, or newly emerging security threats than to uneventful transitions in economic statistics. ⁴ The logic of neorealism would also envision states taking strong action early in their decline, well before an ordinal power transition, to avert this transition in the first place. ⁵ This is important because MacDonald and Parent contend that their study demonstrates the prompt responsiveness of policymakers to structural constraints (p. 21). When the authors make an argument about the timing of a state's response to external stimuli, their coding of the onset of decline becomes immensely consequential. The authors ignore the period leading up to the power transition during which the incentives for retrenchment are equally, if not more, potent. We thus cannot determine whether the declining state's retrenchment policy was a prompt response to an ordinal power transition or a delayed reaction to the gradual and protracted decline that preceded the transition. MacDonald and Parent must more effectively demonstrate that policymakers recognized and responded to the precise causal forces specified in their model.

For instance, British policymakers in the period before World War I did not begin to perceive Britain's decline relative to Germany in 1908, as MacDonald and Parent code the case. The initiation of the German naval program and the escalation of the Boer War were far more consequential in engendering British perceptions of decline, which became evident around the turn of the twentieth century. ⁶ The authors' focus on the 1908...