The false peril of great power retrenchment: Types of strategic withdrawals and their consequences

Abstract

According to conventional wisdom, a great power engaging in international retrenchment regularly incurs tremendous costs. Following its withdrawal from a commitment abroad, the argument goes, windows of opportunity emerge that rivals exploit to their benefit, thus imposing significant costs on the retrenching great power. I argue that pundits and policymakers consistently overestimate the dangers associated with strategic withdrawals: great powers can – and in the past frequently have – successfully engaged in international retrenchment without creating opportunities for their rivals to gain significant strategic benefits. To make this case, I develop a new typology of international retrenchment strategies based on the kind and degree of disengagement they entail and demonstrate that most types do not regularly pave the way for rival gains. I support my argument through a series of plausibility probes: the Soviet retrenchment from Romania in the 1950s; the US retrenchment from Korea in the 1970s; and the US retrenchment from Western Europe in the 1990s.

Introduction

According to popular wisdom about international relations (IR), great powers engaging in international retrenchment regularly end up incurring tremendous costs.¹ As a great power strategically withdraws from some of its commitments abroad, it opens windows of opportunity for others to claim control over the spaces it leaves behind, thus allowing its adversaries to reap important strategic benefits. The costs resulting from the fact that retrenchment allows for rival gains are supposed to be so significant that they usually outweigh whatever benefits a great power might acquire through retrenchment.

Today, this type of argument plays a central role in policy debates on whether the United States should maintain its current position in the Middle East or withdraw from the region to free up resources needed at home or to address security threats in other parts of the globe.² For example, President Biden proclaimed in July 2022 that the United States would not unravel its position of great power supremacy in the Middle East: ‘we will not walk away’, he explained, as this would ‘leave a vacuum to be filled by China, Russia or Iran’.³ Similarly, Gen. Frank McKenzie, commander of US Central Command, told reporters in May 2021 that ‘the Middle East writ broadly is an area of intense competition between the great powers. And I think that as we adjust our posture in the region, Russia and China will be looking very closely to see if a vacuum opens that they can exploit.’⁴

Pundits have expressed similar concerns about the costs US retrenchment would entail by allowing for significant rival gains. Robert Kaplan, for instance, cautions that if the United States were to ‘flinch from its international responsibilities’ and ‘lessen … [its] engagement with the world’, this might hurt US power and could have disastrous consequences ‘for humanity’ overall. An American retreat from ‘a world of rising and sometimes hostile powers’ would not only precipitate the collapse of international order but ‘encourage … bullies’ to extend their influence and thus shift the balance of power in the favor of US adversaries.⁵ Robert Kagan opposes an ‘American retreat’ from abroad for similar reasons. If the United States were ‘to pull back from overseas commitments’, he warns, this would not only ‘worr[y] allies’ but ‘cheer … and embolden … potential adversaries’.⁶

Finally, warnings about retrenchment’s tendency to invite rival gains continue to represent a key reason why many security studies scholars continue to support a US grand strategy of primacy and oppose significant reductions in the United States’ international footprint. In their eyes, an extensive US presence abroad serves to ‘deter states … from contemplating expansion’, meaning that retrenchment would threaten to invite revisionist powers to expand their influence or even contemplate ‘runs at regional hegemony’.⁷ These pessimistic views have served as the basis of US grand strategy during most of the unipolar era and continue to enjoy widespread acceptance among the establishment in Washington today.⁸

In this piece, I posit that these fears about retrenchment’s consequences are overblown because pundits and policymakers consistently overestimate the costs associated with strategic withdrawals: great powers can – and in the past frequently have – successfully engaged in international retrenchment without creating opportunities for their rivals to gain significant strategic benefits.

To make this case, I begin by developing a new conceptual framework which highlights that instances of strategic withdrawal frequently vary along two dimensions, namely, the kind and degree of disengagement they entail. In terms of kind, some forms of retrenchments entail a reduction in a great power’s military footprint only, while others simultaneously work to reduce a great power’s involvement in the area’s economic and diplomatic affairs. In terms of degree, some forms of retrenchment entail only a partial reduction in a great power’s footprint within the respective space, while others are more far-reaching and completely eliminate it. Through their interaction, variation along these two dimensions ultimately produces four ideal types of strategic withdrawals: ‘cutting back’, ‘downsizing’, ‘delegating’, and ‘abandoning’.

I then turn to argue that the conventional wisdom about retrenchment’s consequences – that it is immensely costly because it allows for significant rival gains – rests on a misguided understanding of the conditions which produce windows of opportunity rivals might capitalise on. Pundits and policymakers frequently suggest that whenever a great power withdraws its military capabilities from a certain geographic space, this will allow rivals to expand their influence into the area. Against this, I argue that it is only when a great power loses authority over a given space that such windows of opportunity occur. Since great powers can claim authority abroad without boots on the ground, this means that even when they completely withdraw their troops this need not necessarily invite outside competition for control. In short, by focusing on the withdrawal of military power instead of the collapse of authority, previous work has painted a distorted picture of when and why strategic withdrawals pave the way for significant rival gains.

Taken together, my new typology of retrenchment strategies and the focus on authority collapse suggest that observers regularly overestimate the costs retrenchment entails: of the four types of retrenchment my framework identifies, three – cutting back, delegating, and downsizing – do not regularly create opportunities for rivals to expand their influence. To support this argument, I provide a series of qualitative plausibility probes: the Soviet retrenchment from Romania in the 1950s; the US retrenchment from Korea in the 1970s; and the US retrenchment from Western Europe in the 1990s.

By demonstrating that marked rival gains after great power retrenchment should be a much rarer phenomenon than commonly believed, this piece joins recent efforts to repair retrenchment’s tarnished reputation.⁹ In doing so, it strengthens the case of those who call for the United States to adopt a grand strategy of restraint. Contrary to proponents of primacy, so-called restrainers believe that retrenchment’s benefits usually outweigh its costs and therefore support a marked reduction in the United States’ military footprint abroad.¹⁰ The findings in this piece lend further credibility to their claims: because they falsely expect rivals to regularly gain crucial strategic benefits in the wake of retrenchment, pundits and policymakers commonly overestimate retrenchment’s costs.

The remainder of this piece proceeds as follows. I begin by reviewing and synthesising the IR literature on retrenchment in great power politics, demonstrating that the concept has been employed in a very expansive manner. I posit that there is much to be gained through conceptual disaggregation and proceed to develop my updated framework, which distinguishes between four ideal types of international retrenchment strategies. Thereafter, I turn to reassessing the international consequences of these various strategies, highlighting that most do not present rivals with windows of opportunity to acquire marked strategic gains. To bolster my argument, I provide a series of plausibility probes. Finally, the conclusion recapitulates my main findings, spells out their policy implications, and highlights promising avenues for future research.

Retrenchment: One term, many strategies

Although a staple in the foreign policy and grand strategy discourse for many years, surprisingly few works have provided an in-depth analyses of what exactly implementing a strategy of retrenchment entails.¹¹ In this section, I review current use of the concept, demonstrating that most IR scholars have come to understand ‘retrenchment’ in an expansive manner: within this one category, they group together a variety of disparate strategies great powers might pursue in pursuit of their foreign policy goals.¹²

Consider, for instance, how MacDonald and Parent introduce retrenchment in what arguably constitutes the most thorough discussion of the concept in the literature to date. According to them, ‘at base, retrenchment is an intentional reduction in the overall cost of a state’s foreign policy’ and – as its defining characteristic – thus denotes decreasing grand strategic ambition in one form or another. They then proceed to argue that, in practice, retrenchment can take various forms, ranging from ‘economiz[ing] expanses by cutting … military spending and personnel’, through ‘demoting the importance of some issues’ to ‘renounc[ing] existing commitments altogether’.¹³

In thus grouping a variety of disparate strategies – some far-reaching and costly, others of limited impact and comparatively cheap – they are not alone. For instance, Brooks, Ikenberry, and Wohlforth similarly conceive of retrenchment as a very broad conceptual category. To them, a state engaging in retrenchment can pursue a variety of divergent strategies: it could ‘scale back its global commitments’, ‘curtail or eliminate its overseas military presence’, ‘eliminate or dramatically reduce its global security commitments’, or ‘minimize or eschew efforts to foster and lead … [an international] order’.¹⁴ For Trubowitz, retrenchment is a ‘status quo strategy’ which ‘involves scaling back existing commitments or military capabilities, or both’ – all three of which, of course, would have vastly different international effects.¹⁵ Finally, Atkinson and Williford argue that great power retrenchment denotes ‘declining military expenditures’ but intentionally remain agnostic about whether this means ‘drawing down foreign commitments or decreasing military investment at home’.¹⁶

This is not to say that no efforts have occurred to disentangle some of the conceptual ambiguity surrounding retrenchment. Two efforts stand out as particularly important in this context. First, MacDonald and Parent themselves have proposed the distinction between ‘domestic’ and ‘international’ forms of retrenchment: domestic retrenchment policies, they argue, are inward-looking and serve to ‘reduce military spending, revise force structure, and reform outmoded institutions, while reallocating resources toward more productive investments’; international retrenchment policies, meanwhile, are outward-looking and implemented ‘to defuse potential flashpoints, revamp global posture, and shift burden to allies’.¹⁷ Second, Haynes has rightly critiqued the literature’s ‘unidimensional’ conceptualisation of retrenchment, arguing that this hinders productive theorising about when and how declining states decide to retrench. In particular, he highlights that there exists important variation in the ‘timing and pace’ of strategic withdrawals.¹⁸

Yet, while these represent important first steps in sweeping the conceptual minefield surrounding retrenchment as a great power strategy, I argue that there remains a need for further conceptual disaggregation, specifically vis-à-vis the category of international – or external, outward-looking – retrenchment. To say that a state has, will, or should engage in international retrenchment can mean a variety of different things. At one extreme, it could mean nothing more than that a state is seeking to reduce its international footprint by reducing its military presence in a particular region of the world while maintaining a series of bases and security commitments. On the other, it could mean that it completely abrogates its commitments within a particular area and fully withdraws itself militarily as well as economically and diplomatically. The current, expansive treatment of the term in academic and policy discourses thus hinders a productive discourse on the consequences of engaging in retrenchment: the costs and benefits associated with various forms of retrenchment vary drastically, and to speak of the costs and benefits of international retrenchment overall thus makes little sense.

Political methodologists have long warned that ‘ignoring causally important differences between cases by lumping them together using excessively broad definitions of concepts’ can stand in the way of productive theorising. To gain reliable insights into the causes and effects of a phenomenon that occurs in various forms with ‘causally important differences’ – as is the case when it comes to international retrenchment – they advise ‘disaggregating the overarching concept into several sub-types and engaging in “typological theorization”’.¹⁹

To properly assess the consequences of international retrenchment, and specifically to evaluate the popular belief that a retrenching great power usually allows its rivals to gain marked strategic benefits and thus incurs significant costs, it therefore seems appropriate to employ a more fine-grained conceptual framework that explicitly distinguishes between different types of international retrenchment and thus allows for the kind of type-specific causal theorisation political methodologists recommend in cases like this.

A new typology of international retrenchment strategies

I propose a new conceptual distinction between four types of international retrenchment strategies: ‘cutting back’, ‘downsizing’, ‘delegating’, and ‘abandoning’. What unites the four is they each represent a systematic reduction of a great power’s politico-military footprint in a particular geographic space outside of its homeland, i.e. (international) retrenchment.²⁰ Each strategy differs from the others, however, in terms of the kind and degree of disengagement it entails.

First, in terms of kind, some forms of retrenchments entail a reduction in a great power’s military footprint only, whereas others simultaneously work to reduce a great power’s involvement in the area’s diplomatic and economic affairs. While the former usually takes the form of troop withdrawals, the latter might take a variety of forms. These include but are not limited to: reductions in the economic and military aid previously provided to incentivise friendly alignment; the abrogation of economic or security treaties meant to enshrine local actors’ dependence; and other forms of direct as well as indirect support for the development of greater independence and autonomy in the making of economic and security policy.

Second, in terms of degree, some forms of retrenchment entail only a partial reduction in a great power’s footprint within the respective space, while others are more far-reaching and completely abrogate a great power’s presence. Evidently, this distinction is crucial. For instance, whether a great power fully withdraws military or leaves parts of its force behind must be expected to fundamentally alter the strategic calculus of both local actors as well as third parties.

Through their interaction, these two dimensions produce a four-way distinction between different types of retrenchment (see Figure 1). Before discussing each type in turn, two clarifying notes about this framework are in order. First, note that each form of retrenchment the framework identifies entails at least a partial reduction in a great power’s military presence. Consequently, instances in which a great power moves to curtail its economic or diplomatic involvement in a particular area but leaves its military presence untouched do not qualify as cases of retrenchment for the present purposes.

Figure 1. Four types of international retrenchment.

Second, note that the four ideal types are not meant to capture every possible form of retrenchment great powers ever have pursued or will pursue. Rather, the typology delineates four ideal types which capture distinct patterns of behavior whose differences remain unappreciated by the conventional, broader conception of retrenchment. Future work might find, however, that an even finer-grained distinction could provide for an even more productive discourse and therefore seek to further disaggregate one or more of the types I outline.²¹

Cutting back

The first type of retrenchment great powers can engage in is ‘cutting back’. When cutting back, great powers engage in a partial military withdrawal from a given area. These military adjustments reduce the great power’s military footprint on the ground but are not indicative of the great power’s intention to distance itself from exercising predominant control over the area. Given this type of retrenchment’s decidedly limited character, the hope is that cutting back will not bring about a marked change in existing political circumstances. The great power retains a military presence – at least a token force, but often more than that – and continues to utilise economic and diplomatic means to bolster its influence. In fact, it might up its non-military efforts in order to partially substitute for the reduction in military capabilities it implements. Ultimately, cutting back is a minimally invasive form of retrenchment in that it does not alter a great power’s approach to the respective area: the modalities of control it employs remain the same.

For an illustrative example of this type of retrenchment at work, consider US policy towards East Asia during the Nixon and Carter administrations. During this time period, the United States engaged in a series of significant troop reductions that relocated US troops away from the region in an effort to reduce the costs of the United States’ global posture. Over time, these efforts to cut back reduced the number of US troops in East Asia markedly. For instance, in the mid-1970s over 51,000 troops were stationed in Japan, and over 40,000 could be found on the southern Korean peninsula. By 1992, these numbers had shrunk to fewer than 46,000 and 36,000, respectively. Notably, during this time the United States focused its retrenchment efforts exclusively on reducing its military footprint and did not move to simultaneously curtail its economic and diplomatic efforts in the region. To the contrary, the United States regularly combined the announcement of troop reductions with pledges of additional aid to assuage the leadership of states such as Japan and Korea, which were fearful of a more encompassing US disengagement.²²

Delegating

The second type of retrenchment great powers can engage in is ‘delegating’. When delegating, a great power engages in a complete military withdrawal but seeks to retain its position of predominance over the respective space through non-military tools of statecraft. In essence, this type of retrenchment reflects a great power’s attempt to switch from an arrangement of direct rule, in which boots on the ground are the main way in which it asserts and maintains its influence, to one of indirect rule, in which it retains its influence through local authorities which function as its proxies. Put differently, the desired outcome of delegation as a retrenchment strategy is to bring about the transition to a constellation in which the ‘great power exercises authority not directly, but through national authorities which enter a relationship of subordination with the great power while still retaining their nominal independence’.²³

Haynes has done much to illuminate the strategic logic behind retrenchment as delegation. As he explains in his work on the matter, ‘although declining states often face intense incentives for retrenchment, they retain a vital interest … in the regions currently under their protection’. Hence, they are often reluctant to withdraw unless ‘they expect their preferred order to survive this withdrawal’. This, in turn, is usually only the case if they can identify ‘“successor states” to which the declining state can devolve its … responsibilities’.²⁴

A paradigmatic example of a great power engaging in this type of retrenchment can be found in Soviet policy towards Romania in the late 1950s.²⁵ In the early post-war years, the Soviet Union centrally relied on its troops to exercise its control over Central and Eastern Europe. In countries such as Romania, Soviet troop deployments ‘reached the level of one soldier for every 10 to 20 local persons’. Direct military control served as the foundation of the Soviet Union’s sphere of influence in the area. By 1955, however, Moscow sought a more sustainable imperial arrangement in which it could rule over Romania indirectly rather than through direct military control. Soviet leaders therefore decided to engage in a complete withdrawal of troops from Romania by mid-1957, and the withdrawal was successfully completed by the summer of the following year.²⁶ Moscow continued to exercise control over Romania afterwards but now did so without a military presence within Romania itself.

Downsizing

The third type of retrenchment great powers might pursue is ‘downsizing’. When downsizing, a great power significantly reduces its military as well as economic and diplomatic footprint in an area outside of its homeland. The kind of disengagement it entails thus touches all tools of statecraft: the great power reduces its direct military involvement, e.g. by engaging in troop withdrawals; it reduces its economic involvement, e.g. by significantly limiting aid provisions; and it reduces its diplomatic support, e.g. by granting indigenous political actors greater domestic and foreign policy autonomy. At the same time, the reductions implemented are only partial: for instance, the great power might retain parts of its military presence, provide aid provisions of smaller magnitude, and continue to offer less-encompassing diplomatic support. Great powers that engage in this type of retrenchment thus do not fully disengage from and abolish their involvement in the respective area’s political affairs. At the same time, downsizing does represent a determined effort to reduce its engagement across the board, thus leading to pronounced cuts in the resources it spends through its involvement.

US policy towards Western Europe at the end of the Cold War constitutes an illustrative example of this strategy at work. Given how large the US-American commitment to the area had been during the Cold War, it is little surprising that leaders in Washington sought to reduce the costs of their involvement after the collapse of the Soviet Union: with the demise of its once-formidable peer competitor, a key rationale for the large US politico-military footprint on the ground dissipated.²⁷ The United States thus decided to drastically cut the number of troops stationed on the European continent. In the period between 1986 and 1990, ‘on average, 311,870 [U.S.] troops were stationed in Europe per year’. This number ‘was slashed by two-thirds after the Berlin Wall fell, to an average of 109,452 troops per year during 1996–2000’.²⁸ In addition, US officials moved to reduce their diplomatic and economic involvement by supporting the continuation of European integration processes and the continent’s move towards greater political autonomy. While far from completely disengaging itself either militarily, economically, or diplomatically, in the immediate post–Cold War years the United States thus sought to significantly downsize its commitment to Western Europe compared to what it had been in prior decades.

Abandoning

The fourth and final type of retrenchment great powers can engage in is ‘abandoning’ an area outside of their homeland. Abandonment is the most extreme form of retrenchment the present conceptual framework identifies. As is the case when delegating, a great power that abandons a certain part of the globe engages in a full military withdrawal. In addition, however, it also abrogates all other, non-military sources of influence, including prior economic and diplomatic efforts to maintain its position of predominance and shield the area from outside influence. Abandonment thus represents a great power’s decision that, for whatever strategic reasons, it will no longer maintain its involvement in this part of the world and unilaterally abrogate its position of predominance. Ceasing all prior military, economic, and diplomatic involvement, the great power completely disengages and leaves the area to its own devices.

Britain’s abandonment of some of its peripheral colonial possessions in response to its drastic material decline after World War II represents a prime historical example of this type of retrenchment at work.²⁹ For instance, ‘liquidat[ing] hitherto inviolable traditional commitments’, Britain ‘abandon[ed] India … in haste’ shortly after the war concluded.³⁰ The twin threats of material decline and growing nationalist sentiments in the colony had proved too much to handle for the struggling metropole, causing it to abandon what had once constituted a highly prized colonial possession.³¹ In the course of retrenching from the Indian subcontinent, London not only decided to fully withdraw its troops – the last of which left by 1948 – but also granted the new states of India and Pakistan political independence. Following formal independence, although India remained part of the Commonwealth, ‘India and Britain became … less important to each other in trade, finance, education and defence’, and over the years disengagement became so pronounced that the two countries ‘had weaker links than made sense given the interests of elites or masses in either country’.³²

Reassessing retrenchment’s consequences

With this more refined understanding of retrenchment in hand, we can now turn to reassess its international consequences. As noted in the introduction, both pundits and policymakers have long warned of retrenchment’s immense costs which result from the fact that it opens windows of opportunity for rivals to improve their position in the international system. But does retrenchment, indeed, regularly pave the way for rivals to acquire important strategic assets?

I argue that it does not. To make this point, I proceed as follows. I first show that it is the collapse of a great power’s authority, not the withdrawal of military force per se, that presents rivals with the opportunity to expand their influence. I then make the case that great powers engaging in three of the four types of retrenchment my framework identifies should regularly be able to avoid precipitating such authority collapse. Finally, I highlight that even when great powers engage in the one type of retrenchment that does create such opportunities for rival gains, there is good reason to believe that rivals will frequently not be able to acquire these gains at negligible costs.

What invites rival expansion? Centring great power authority collapse

In their warnings about the dangers of strategic withdrawals, pundits and policymakers centrally build on the assumption that whenever a great power removes some its military capabilities from a certain geographic space, this will open a window of opportunity allowing its rivals to expand their influence into the area.³³ In this section, I challenge this popular belief, arguing that it is only when a great power’s authority collapses that the costs for rivals to extend their influence into the space concerned sink dramatically.

Following Weber, authority denotes the ‘legitimate exercise of imperative control’ and is a function of both material capability and legitimacy.³⁴ The possession of preponderant coercive capabilities and the ability to use them to issue coercive threats is the cornerstone of any form of imperative control: it generates widespread obedience as even those ‘subjects who view the ruler’s orders as fundamentally unjust have an interest in compliance if said orders are backed by preponderant power and the threat of sanctions’. At the same time, stability and longevity of rule require at least some degree of legitimacy in the sense that subjects follow the ruler’s commands quasi-voluntarily, i.e. they willingly comply with its demands for whatever ulterior motivation and therefore do not need to be constantly monitored and punished.³⁵

Historically, great powers have exercised authority not only over their own homeland but also minor powers and other political organisations outside of it.³⁶ Such cases of spatially extensive authority in which great powers establish authority over other polities usually take one of two primary forms. First, great powers can exercise authority through the maintenance of a formal empire. In this constellation, an imperial state or metropole acquires territorial control over political spaces outside of its own homeland and wields authority over them as formal colonies or dependencies. Formal empires thus involve direct rule and formal incorporation into the metropolitan state.³⁷

Second, great powers sometimes exercise authority without acquiring formal territorial control. The political spaces they rule in this manner have variably been called their ‘informal empires’ or ‘spheres of influence’.³⁸ In these cases, a great power exercises authority not through territorial control but through national authorities which enter a relationship of subordination with the great power while still retaining their nominal independence. Just like formal empires, these arrangements of imperial control are dependent upon both pillars of authority, i.e. not only the ability to inspire a sufficient degree of willing obedience but also to project preponderant power into the space concerned. Consider, for instance, the US sphere of influence in Latin America: although the United States’ position of predominance does not require a permanent military presence on the ground as its local subordinates generally comply with its demands, the stability of its rule depends crucially on its ability to enforce its dictates militarily if push came to shove.

When a great power exercises authority over a political space, this effectively limits the ability of other great powers to ‘gain direct access and control of’ the ‘population and territory’ it contains, as rivals could hope to extend their influence into the space only at tremendous costs.³⁹ Seizing authority over another great power’s territorial possessions would be possible only by conquering and occupying them. Attempts to challenge a great power’s control over its sphere of influence are similarly costly. After all, its hold on authority over these spaces implies not only that the great power is able to project preponderant power there, but also that its rule enjoys sufficient legitimacy. Any adversary seeking to assert control in its stead must not only expect to trigger a military confrontation with a peer competitor but also faces the prospect of an indigenous population that is committed to retaining said peer competitor’s rule. In the words of Eric Forman, by claiming authority over various geographic spaces great powers thus erect ‘fences or “hard shells” around specific portions of the globe’, effectively shielding them from the influence of their rivals.⁴⁰

When a great power’s authority over a space collapses, however, this destroys the protective shell around it. Such authority collapse can occur for two main reasons. One possibility is that the great power loses large parts of the material basis of its rule, i.e. it is no longer capable of projecting preponderant material power into the respective space and is therefore no longer in a position to fend off outside challenges to its position of predominance. The other is that the great power experiences a pronounced loss in legitimacy, i.e. it loses its ability to elicit quasi-voluntary compliance with its commands from subordinate political actors. As the former subordinates turn away from the great power, they will not only refuse to defend its rule but might even invite attempts by other powers to oust and replace it in the hope of thus acquiring greater security or other material as well as normative benefits for themselves. Overall, whether the collapse of a great power’s authority is due to the erosion of its ability to project power into the space, its legitimacy, or both, it radically alters third parties’ strategic calculus: specifically, the great power’s peers realise that they now face the unique opportunity to seek and claim control over a space that was previously impervious to their rule.

Shifting our analytic focus from the removal of military capabilities to authority collapse when discussing strategic withdrawals’ likely consequences might at first appear to be nothing more than a minor semantic move. On closer inspection, however, it has far-reaching consequences for how to think about the connection between retrenchment and possible rival gains. Most importantly, it reveals that the withdrawal of military force from a particular space alone is neither necessary nor sufficient to invite rival expansion: it is possible that a great power maintains authority over spaces outside of its homeland even without a direct military presence within them, and it is similarly possible that it loses authority over spaces abroad despite a continued military presence. As I now turn to show, unduly focusing on the removal of military power per se rather than the collapse of authority has thus led pundits and policymakers to reach distorted assessments of international retrenchment’s costs.

Why many strategic withdrawals do not pave the way for rival gains

All types of retrenchment my framework outlines entail at least a partial reduction in a great power’s military presence on the ground. Hence, pundits and policymakers have concluded, retrenchment regularly invites rival expansion. This thinking, however, is too simplistic. As noted above, windows of opportunities rivals might exploit emerge when a great power’s disengagement causes its authority to collapse. In this section, I argue that there are strong reasons to believe that great powers engaging in three out of the four forms of strategic withdrawal my conceptual framework identifies – cutting back, delegating, and downsizing – will often be able to retain their authority over the respective area. Consequently, no opportunities for rival expansion emerge that would pose a threat to the retrenching great power’s security. In other words, these forms of retrenchment do not carry the costs strategic withdrawals are frequently alleged to entail.

Let us, first, consider cutting back, i.e. retrenchment which takes the form of a great power engaging in a partial military withdrawal from an area outside of its homeland. When cutting back, the retrenching great power continues to retain at least a token force – and often more than that – in the area, which helps maintain order on the ground and deters outside challenges to its rule, at a minimum, by functioning as a tripwire.⁴¹ In addition, it continues to bolster its influence through non-military means such as the provision of economic aid or diplomatic assurances, which it might even increase in order to assuage any domestic opposition against the partial troop withdrawal and signal to other parties that it remains committed to the defence of the area. In principle, great powers should thus be able to retain their authority when engaging in this type of retrenchment.

Given the limited scope and depth of disengagement cutting back entails, the risks associated with it are relatively small. Admittedly, it remains a theoretical possibility that a limited troop withdrawal will set off a cascade of events through which the great power’s authority becomes significantly weakened.⁴² For example, domestic as well as international observers could perceive it as the prelude to a more thorough disengagement, conclude that the great power’s commitment to the area is wavering, and thus feel incentivised to challenge its position of predominance. Yet the likelihood for such cascades to occur appears slim, as the threat of great power war continues to loom in the background, and local actors remain aware of the great power’s continued ability to militarily fend of challenges to its rule if necessary. Most of the time, cutting back should thus be possible without worrying about creating opportunities for rivals to expand their influence abroad.

Second, even a complete troop withdrawal – such as occurs when delegating – need not precipitate the collapse of a great power’s authority. As explained above, when a great power decides to retrench by delegating, it essentially attempts to switch from a political arrangement in which boots on the ground are a key foundation of its control over the area to one of indirect rule, i.e. in which it continues to exercise predominant influence in the respective space but does so remotely, usually by recruiting local political authorities as its proxies. The underlying logic of delegation – and what makes it so appealing for great powers in the first place – is that it allows them to retain predominant influence while dramatically reducing the costs of doing so by eliminating the direct military presence on the ground. The goal is not to rescind one’s authority, but to exercise it through an alternative, less costly arrangement. Put differently, pursuing delegation is centrally motivated by the goal of withdrawing military forces without precipitating the kind of authority collapse rivals might be able to exploit.

In principle, delegation of course carries the risk that the switch to indirect rule fails. Hoping to exercise the same degree of imperative control as before, the great power might find that it overestimated its ability to rule from afar. For instance, local political actors might turn out to be unreliable either in terms of their will to behave in accordance with the great power’s dictates or in terms of their ability to maintain military capabilities sufficient to deter outside intervention. Yet, as the strategic actors that they are, great powers are well aware of this possibility and will therefore only pursue delegation when they are near-certain that it will succeed, i.e. they are confident in their prospective proxy’s reliability. What should thus occur more frequently than great powers pursuing delegation and later finding that it failed is great powers planning to pursue it, concluding that it is not viable, and therefore never implementing it. Not coincidentally, there are few examples in modern history in which great powers pursued delegation and failed.⁴³

As an illustrative example, consider US plans for a full military disengagement from Western and Central Europe after the conclusion of World War II. It is increasingly well documented that, from the United States’ entry into the war onwards, the leadership in Washington hoped to avoid any prolonged military entanglement on the continent.⁴⁴ During the war, Franklin D. Roosevelt declared that ‘the United States will be only too glad to retire all its military forces from Europe as soon as this is feasible’, and throughout the two following administrations US officials did not give up on the idea that any direct military involvement on the continent should not be allowed to become a perpetual one.⁴⁵ Yet the United States never actually implemented a full military withdrawal and – instead of pursuing retrenchment as delegation, as initially planned – US troops remained on the continent indefinitely after the conclusion of the war. The key underlying reason for this was that the power differential between the Soviet Union and the Western European states, despite the profound aid they had received through the European Recovery Program and other initiatives, remained too stark as Soviet capabilities soared. From Washington’s perspective, retrenching by delegating was simply not feasible, and they therefore never engaged in this type of strategic retrenchment.⁴⁶

Finally, consider downsizing. Recall that when retrenching by downsizing a great power initiates a partial military withdrawal and simultaneously works to reduce its politico-economic involvement in an area outside of its homeland. The type of disengagement this form of retrenchment entails is encompassing in scope – it targets both military and non-military foundations of influence – but limited in degree, as the great power neither fully withdraws its troop presence nor completely abolishes its economic and diplomatic involvement. Whether this type of retrenchment will pave the way for rival gains will vary from case to case: it centrally depends on how large the retrenching great power’s remaining politico-military footprint is after the completion of the downsizing process.

Yet, given that it retains a military presence and simultaneously continues to utilise other tools of statecraft (albeit to a lesser extent) to bolster its relationship with local political actors, in many cases the retrenching great power’s authority will likely be somewhat weakened but not erode to the point that outside actors would feel compelled to compete for control over the area. The retrenching great power’s remaining forces will help deter outside aggression, and the economic and diplomatic support it continues to offer not only signals its continued interest in shaping the area’s political future but will help bolster indigenous forces and secure their continued allegiance. In consequence, while the reductions downsizing entails might, in some cases, indeed end up being so pronounced that they precipitate a collapse of authority, there are good reasons to believe that there will exist just as many cases in which great powers downsize without opening windows of opportunity for their rivals to exploit.

The consequences of abandonment: Strategic disaster guaranteed?

But what about the fourth and most extreme type of retrenchment, i.e. abandoning? When a great power decides to abandon a certain part of the world, it effectively revokes its authority over it. Its military forces are fully withdrawn, it no longer offers economic or security benefits to incentivise subordination from local political actors, and thus effectively abolishes the hierarchical relationship that served as the foundation of its influence. Given the all-encompassing scope and depth of this type of retrenchment, it is essentially unavoidable that it opens a window of opportunity for other powers to compete for authority over the respective space. As opposed to those cases in which the other three types of retrenchment are concerned, prima facie it therefore appears more plausible to argue that when a great power retrenches by abandoning it is bound to incur significant costs.

Still, one should be careful not to overestimate the risks associated with abandonment either. While the collapse of the retrenching great power’s authority will noticeably reduce the costs for its rivals to try and claim control over the space it leaves behind – most importantly, they must no longer fear a direct military confrontation with a peer competitor – this does not necessarily entail that said costs will be sufficiently low to render expansion attractive. When faced with the threat of a great power seeking to claim control over them, historically, nation-states have responded not by appeasing or bandwagoning with the outside power but by balancing against it.⁴⁷ Because of these local balancing dynamics, other great powers seeking to expand their influence after the retrenching great power’s departure are thus still likely to confront a respectable obstacle that could prove formidable enough to deter them from realising whatever expansionist ambitions they might harbour.⁴⁸

Perhaps even more importantly, one should be careful not to take the above discussion to suggest that abandonment will necessarily represent a strategic disaster for the retrenching great power. This piece is primarily concerned with the costs of retrenchment and how they have been consistently overestimated. Whether retrenchment in any given instance leads to increases or decreases in a great power’s security, however, depends as much on the benefits it acquires by pursuing this course of action. In other words, even if retrenchment by abandonment does open a window of opportunity that is successfully exploited by rivals which acquire important strategic assets, these costs – significant as they may be – might be outweighed by the benefits the retrenching great power acquires. The latter can include, among others: a lower likelihood of involving the great power’s troops in active combat; considerably less strain on the great power’s economy, on whose health its power crucially depends in the long run; the ability to invest more resources to address domestic issues of concern; and freeing resources to deal with more vital security threat in other regions of the world.⁴⁹ Indeed, being the strategic actors that they are, we should not observe great powers abandoning certain areas of the globe unless they are confident that the resultant benefits will ultimately outweigh the costs they will incur by engaging in this form of retrenchment.⁵⁰

Stories of success: Great power retrenchment in action

Let me now turn to the historical record to gather empirical support for my argument that most forms of retrenchment – cutting back, delegating, and downsizing – do not regularly create opportunities for rivals to expand their influence at significantly lower costs. To show that such cases of successful retrenchment are more than just a theoretical possibility, I provide three plausibility probes for my argument, each of which concerns a different type of retrenchment. First, I investigate the United States’ retrenchment from Korea during the 1970s – a case of cutting back – and show that the troop reductions Nixon implemented did not undermine US authority over the southern peninsula. Second, I assess the Soviet Union’s retrenchment from Romania during the 1950s – a case of delegating – and find that even though Moscow engaged in a complete troop withdrawal, this did not undermine its authority. Finally, I take a look at the United States’ retrenchment from Western Europe in the immediate post–Cold War years – a case of downsizing – and show that it weakened but ultimately did not erode Washington’s position of authority vis-à-vis the region.

Before moving on, a brief note about my empirical approach – that is, utilising a small number of short case studies as plausibility probes – is in order. After all, as Harry Eckstein notes, ‘it is true that case studies have been little used in political studies as plausibility probes’, which ‘is largely due to the fact that the idea of any sort of plausibility probe is foreign to the field’. The logic behind this approach is that ‘after hypotheses are formulated … [a] stage of inquiry preliminary to testing sometimes intervenes and ought to do so far more often than it actually does in political study’, namely, ‘probing the “plausibility” of candidate-theories’. Note that ‘plausibility here means something more than a belief in potential validity plan and simple’, yet ‘also means something less than actual validity, for which rigorous testing is required’. Essentially, plausibility probes allow scholars to develop ‘a reasoned, not merely intuitive, “feel” for the odds [for or] against a theory’, help ‘establish that a theoretical construct is worth considering’, and serve as ‘suggestive tests … before more rigorous tests are conducted’.⁵¹

Given that the present work’s focus lies on proposing a new theoretical intervention to cast doubt on a widely held conventional wisdom, it falls beyond its scope to also develop, conduct, and analyse the kind of extensive empirical tests necessary to reach a final decree about the contending views’ validity. At the same time, if this piece were to provide no empirical evidence at all, its central proposition would rightly be rejected as not worthy of serious consideration. Presenting a number of plausibility probes thus represents an appropriate ‘middle way’, and the success of recent IR studies with similar epistemic aims in utilising this approach should increase our confidence in its effectiveness.⁵²

Cutting back: The US retrenchment from Korea, 1970s

Facing the need to drastically cut the US defence budget, which due to the Vietnam War had risen to $81 billion in 1969, President Nixon decided to retrench from Korea during the 1970s.⁵³ Although initially hoping that it might be possible to engage in a full military withdrawal, the United States quickly realised that this was not feasible at the moment, even if existing treaty commitments were to be maintained after the military departure. A key reason for this were continued South Korean pleas that indigenous capabilities to withstand an attack from the North were far from sufficient. Accordingly, at this time retrenchment from Korea remained decidedly limited in both kind and degree.

In the terminology of the framework developed here, it represented a case of cutting back. Nixon’s focus lay exclusively on reducing the US military footprint on the ground, leaving other sources of influence not only intact but in fact bolstering them, e.g. by offering additional security assurances and promises of extensive aid. Although the United States decided in the spring of 1970 to withdraw roughly 20,000 troops from Korea, it provided President Park’s government with a series of reassurances in return. These included ‘a pledge not to withdraw any more troops’, a ‘written guarantee that the United States would support South Korea if the North attacked’, as well as the promise of ‘an additional $3 billion worth of military assistance between 1970 and 1975’.⁵⁴ In addition, the United States far from eliminated its military presence on the ground in Korea. To the contrary, its remaining troop presence continued to play an integral part in its plan to maintain influence over the southern peninsula.

Precisely because the United States maintained a significant military presence even after cutting back and, in addition, increased its use of non-military tools of statecraft to bolster its relationship with the South Korean government and the latter’s ability to provide for its security, the United States’ decision to cut back from Korea during this time did not undermine its authority. In consequence, other powers – and most importantly Soviet leaders – did not view Washington’s retrenchment as opening an opportunity for them to extend their influence into the area. To the contrary, for observers in Moscow and beyond, US retrenchment at this time did not bring about a meaningful change in political conditions. In fact, they essentially considered it a non-issue. It was common knowledge that South Korea continued to fall squarely under the hegemony of the United States, in both the security and economic realms.⁵⁵

In short, after Nixon’s decision to cut back from Korea, there was never any serious worry – or, from the perspective of US rivals, hope – that this might precipitate the erosion of its position of dominance vis-à-vis South Korea. No opportunities for rival gains emerged, as US authority remained soundly intact.

Delegating: The Soviet retrenchment from Romania, 1950s

The previously discussed example of a great power retrenching by delegating – the Soviet Union’s retrenchment from Romania in the 1950s – strikingly demonstrates that this type of retrenchment need not create opportunities for rival gains either. Recall that, following World War II, the Soviet Union centrally relied on a massive troop presence on the ground to both assert its control over Eastern Europe and prevent an extension of US influence into the region. By the mid-1950s, however, Khrushchev began to ‘seek a more stable and permanent basis of relations with its satellites’. Crucially, ‘this did not mean willingly losing control over them. Instead, it meant making certain readjustments in order to retain a maximum degree of influence and control, given the changing circumstances since Stalin’s death.’⁵⁶

On all counts, these efforts to retrench by delegating without seriously undermining Moscow’s position of predominance vis-à-vis Romania proved successful. Despite engaging in a full military withdrawal, the Soviet Union was able to retain Romania as part of its sphere of influence and avoided leaving a window of opportunity for the West to compete for predominance in this part of the world. US leaders concluded from the get-go that the Soviet Union’s complete military withdrawal from the area did not seriously alter the political circumstances in Eastern Europe. It had done nothing to undermine the Soviet Union’s authority in the region: Romania still fell squarely within the Soviet Union’s sphere of influence. Accordingly, ‘in Washington the withdrawal decision was not thought to be an important Soviet move’. As several US officials – including the assistant secretary of state for European affairs, the assistant secretary of state for policy planning, and the special assistant for Soviet bloc political-military affairs at the time – recounted afterwards, ‘the Soviet troop withdrawal was seen as a minor action with no relevant impact’.⁵⁷

The main reason for why Western observers considered the Soviet Union’s retrenchment from Romania to be geopolitically insignificant is that it did nothing to undermine Moscow’s position of authority. In the Romanian government, the Soviet Union had identified a sufficiently reliable proxy that would willingly comply with its demands. Thanks to its geographic proximity, it further retained the ability to project power into Romania even without a direct military presence and could thus ensure the stability of its influence. Consequently, no window of opportunity emerged that would have invited competition for control over the area.

Even before the Soviet Union withdrew its troops, Western intelligence analyses explained along these lines that a ‘Soviet withdrawal alone would be less important than a settlement which allowed the satellites to pursue at least a neutral policy’.⁵⁸ Put differently, even a complete military withdrawal would be meaningless since it would do little to undermine the Soviet Union’s ability to control political developments on the ground. Tellingly, the majority of Romanians themselves shared this assessment, ‘perceiv[ing] the development as one of only limited significance’. Diplomatic reports from Bucharest spoke of the population’s ‘cynicism’, ‘skepticism’, and ‘apathy’, indicating that few expected the Russian troop withdrawal to meaningfully alter political realities on the ground.⁵⁹

In short, all observers – whether in the Soviet Union, the United States, or Romania itself – concurred that the Soviet retrenchment from Romania did not meaningfully alter the political landscape in Eastern Europe: Soviet authority over Romania remained intact, and no opportunity emerged for the West to expand its influence into the region.

Downsizing: The US retrenchment from Western Europe, 1990s

Finally, let us turn to a case of downsizing, namely, US efforts to reduce their commitments to Western Europe in the immediate post–Cold War years. As previously noted, following the collapse of the Soviet Union, Washington sought to notably reduce its politico-military footprint on the continent by engaging in significant troop reductions and supporting the continent’s move towards greater political autonomy.⁶⁰

As Geir Lundestad explains, several reasons caused the George H. W. Bush administration to embark on this course. For instance, ‘the liberation of Eastern Europe had to be buttressed financially and commercially and, with Reagan having run up huge budget and trade deficits, Washington now left much of the initiative for economic dealings with Eastern Europe to the EC [European Community]’. Even more importantly, further integration of reunited Germany ‘was seen as essential for continued stability’. Finally, the ‘EC was also valuable in the perspective of the American desire for burden-sharing … If Washington was to do less in Europe and more in other parts of the world, the European capitals had to do more in Europe.’⁶¹

To a notable degree, the decision to grant Europe greater politico-economic autonomy in the immediate post–Cold War years paved the way for the emergence of Europe as a political power. As Moravcsik summarises, by 2010 European integration had ‘enjoyed an astonishingly successful run’. Among other achievements, ‘it completed the single market; established a single currency; created a zone without internal frontiers (“Schengen”); [and] launched common defense, foreign, and internal security policies’. Consequently, the European Union emerged ‘as a major global force’ to be reckoned with.⁶² In the words of David Lake, as a result of the increasing autonomy Western Europe gained in both the economic and security realm, ‘America’s hierarchies in Europe were clearly waning’ following the end of the Cold War.⁶³

However, although the United States’ new politico-economic role in Western Europe was thus notably limited relative to the one it had played during the Cold War, it would be fundamentally mistaken to conclude that these developments caused US authority over Western Europe to erode. While the United States supported the idea of granting Western Europe an increasingly independent economic and political role, this did not mean that it sought to grant full foreign policy autonomy. To the contrary, ‘in Washington the assumption was still that the United States would act as the undisputed leader, despite the relative rise of Western Europe’.⁶⁴ Both militarily and economically, Western Europe continued to be dependent on close relationships with the United States. Although the number of US troops stationed in Europe was cut significantly at the beginning of the unipolar era, the US military presence in Western Europe remained one of the largest in the world, exceeding 100,000 until well after the year 2000.⁶⁵ US-led NATO continued to play an integral role in Europe’s security structure, and the security guarantee built into it continues to represent one of the main reasons why Western European nations appear to see little need to increase their own defence efforts even today.

For well beyond the post-1990 decade, Washington has thus been able to exert tremendous influence on the behaviour of Western European states. Rather than allowing the order which had served to facilitate cooperation in the West during the Cold War to collapse, US leaders worked hard to not only maintain but expand it in the years after 1990. Although the establishment of the ‘new world order’, as President Bush came to call it, was centrally inspired by ideological concerns and motivations, it also represented an institutional arrangement which would serve to lock in US dominance vis-à-vis Europe and beyond.⁶⁶ This was no coincidence. To the contrary, the United States consciously moved to maintain its hegemony over Western Europe even after the fall of the Berlin Wall.⁶⁷

Given the remarkable success of these efforts, US authority remained resolutely intact as the world entered the era of unipolarity. Consequently, no window of opportunity ever materialised that would have allowed either one of the European states or an outside power to seriously challenge US dominance on the continent. The United States consciously and successfully maintained its position of authority and thus Europe’s continued pro-US alignment.

Conclusion

It would be an understatement to say that retrenchment has a bad reputation. Many pundits and policymakers today believe that great powers which withdraw from commitments abroad ultimately end up incurring significant costs. Specifically, they argue that retrenchment opens windows of opportunity rivals can exploit to expand their influence and acquire important strategic assets, thus shifting the balance of power in their favour and increasing the retrenching great power’s vulnerability.

In this piece, I have argued that, while intuitively appealing, these arguments paint a distorted picture of retrenchment’s consequences, as they consistently overestimate its costs: they fail to appreciate that retrenchment can take various forms, many of which do not regularly open windows of opportunity for rivals to capitalise on. In short, the causal relationship purportedly linking retrenchment to significant rival gains does not operate as commonly postulated.

These findings have important implications for policymakers today. In particular, they speak to the current debate surrounding US grand strategy and the question of whether it would be wise for the United States to retrench from some of its commitments abroad. This piece suggests that it is likely possible for the United States to engage in some types of strategic withdrawals without allowing US rivals such as China or Russia to acquire substantial geopolitical benefits. In this vein, the three cases of successful retrenchment discussed in this piece provide important insights into why and how the great powers at the time were able to avoid eroding their authority while retrenching and thus provide a series of valuable lessons for policymakers today. Three stand out in particular.

First, whenever the troop reductions implemented were only partial, this served the retrenching great power well: whether in the case of its withdrawal from Korea during the 1970s or from Western Europe in the 1990s, the fact that the United States continued to retain some of its troops on the ground helped ensure that its position of authority would persist. Said troops not only served to remind the local populations of the United States’ special interest in – and ability to shape – political developments on the ground, but more importantly deterred outside powers from challenging US predominance.

Second, as the case of the Soviet Union’s retrenchment from Romania shows, as long as the retrenching great power retains the ability to project preponderant military force into the area it leaves, even a full military withdrawal is unlikely to cause the erosion of its authority. Local political actors will realise that – even in the absence of a direct troop presence – the threat of force always looms in the background. Put drastically: misbehave, and foreign troops might return. Similarly, third parties that might consider competing for authority after the troop withdrawal has concluded will be aware of the retrenched great power’s ability to come to the area’s defence if needed.

Finally, as illustrated neatly by the Korean case, increases in economic and diplomatic concessions can go a long way towards compensating for troop reductions. As much as Nixon’s decision to reduce the number of stationed US troops caused anxiety among the South Koreans as to whether they would be able to rely on the United States going forward, his decision to combine these cutbacks with new allocations of aid, a commitment to not withdraw any more troops, as well as a promise to come to South Korea’s aid in the case of an attack from the North, did much to alleviate said anxiety and thus bolstered the US position post-retrenchment.

Systematically investigating the success conditions for the various forms of retrenchment beyond the three cases discussed here represents one of several exciting avenues for future research opened by the present piece. In addition, scholars might seek to investigate whether theoretical and policy discussions would benefit from an even finer-grained conceptual distinction regarding retrenchment than the one advanced here.

Overall, my hope is that, by providing a novel conceptual framework distinguishing between various types of international retrenchment, this piece will enable better causal theorising and thus a more productive debate about the costs and benefits of great power retrenchment. Only once our discussions of retrenchment move past simplistic causal associations based on crude conceptual foundations can we hope to generate solutions to the grand strategic challenges confronting policymakers in the present moment.