Key Takeaways
Treat covert regime change as a tempting shortcut with a poor record
A routine instrument. During the Cold War, the United States attempted 64 covert regime changes and only six overt ones. The appeal was secrecy: local actors would do the visible work, while Washington hoped to deny responsibility and avoid retaliation, expense, and public criticism. The numbers expose the gap between promise and performance. Only 25 covert campaigns installed a US-backed government, a success rate of 39 percent.
Low visibility, real consequences. Many operations were exposed or suspected, and success at removing a leader did not guarantee a durable ally. The record also includes civil conflict, government repression, and worsening relations with the target. Covert action could make intervention seem politically affordable to decision-makers without making its consequences small.
What's striking is how secrecy can distort a decision before an operation even begins. When leaders expect deniability, they may treat a high-risk plan as if its downside were limited. That resembles a familiar problem in organizational governance: hidden decisions weaken scrutiny while encouraging excessive confidence. The book's Cold War dataset makes the pattern concrete, though it cannot establish that every covert intervention in every era performs this badly. Its strongest warning is narrower and more useful: concealment lowers the apparent price of intervention, not necessarily the actual price.
Intervene only when interests are irreconcilable and a credible successor exists
Two conditions matter. O'Rourke argues that regime change becomes attractive when a dispute reflects a chronic clash of security interests, and ordinary bargaining or pressure seems unable to resolve it. A target leader may refuse demands because compliance could threaten the regime's survival or the country's security. The second requirement is a plausible replacement: someone able to take power, govern, and pursue policies Washington prefers.
A replacement must have roots. In the Dominican Republic, US officials hesitated to remove Rafael Trujillo until they believed moderate opponents could take over. In South Vietnam, American leaders debated a coup against Ngo Dinh Diem but repeatedly found no reliable alternative. A potential successor who shares the intervener's preferences on paper is not enough if that person lacks domestic support or governing capacity.
This is a useful corrective to the idea that military or economic superiority makes political engineering easy. Intervention depends on local coalitions, not just external resources. The argument also connects to bargaining theory: when a government sees compliance as a threat to its own survival, pressure can harden resistance rather than produce agreement. Yet the test of a credible alternative is partly subjective. Officials may mistake access to a willing faction for evidence that it can govern, or confuse alignment with Washington for legitimacy among citizens.
Classify the motive as rollback, containment, or regional control
Three security logics. The book groups US campaigns into three ideal types. Offensive operations, called rollback during the Cold War, sought to weaken a military rival or fracture its alliance. Preventive operations, associated with containment, aimed to stop a state from joining a rival camp or becoming a future threat. Hegemonic operations sought to preserve a hierarchy in which neighboring states remained politically subordinate to Washington.
The map clarifies motive. Operations behind the Iron Curtain fit rollback. Efforts in places such as Iran or Vietnam fit containment. Interventions in the Dominican Republic reflected a longer US effort to maintain regional dominance in the Western Hemisphere, a goal that predated the Cold War. These categories can overlap, but they help distinguish a direct military threat from anxiety about future alignment or loss of regional authority.
The typology is valuable because it separates motives that are often collapsed into a single story about fighting communism. The same anti-communist language could describe fear of a military rival, prevention of a future alliance shift, or defense of US dominance over nearby states. Those are different strategic problems, with different risks and likely responses. The categories are analytical tools rather than watertight boxes, however. Policymakers can hold several motives at once, and a tidy classification may make a contested or opportunistic decision look more coherent than it was.
Secrecy lowers exposure costs while weakening execution and oversight
The secrecy trade-off. Covert action can reduce the chance of military retaliation, lower direct costs, and shield leaders from criticism for violating public commitments. But maintaining deniability limits the size of an operation. The larger the effort, the more people know about it, and the harder it is to keep hidden. That constraint matters most when opposition forces are too weak to overthrow a powerful government without substantial outside help.
Hidden plans make poor plans. Secrecy can also narrow who reviews a proposal, blur instructions as they pass through agencies, and encourage weak contingency planning. The early US attempts to infiltrate Albania and Soviet Ukraine illustrate the result: local agents were captured or killed, and security services often appeared to anticipate their arrival. In Vietnam, US efforts to build resistance in the North likewise failed to create an effective movement.
The tension resembles a principal-agent problem: leaders delegate covert tasks to intermediaries, then lose visibility into how those tasks are interpreted and carried out. Secrecy protects a decision from public scrutiny, but it can also protect it from useful internal challenge. This helps explain why a government might persist with a plan despite poor evidence of feasibility. Still, secrecy is not the only cause of failure. The strength of the target state's security apparatus and the weakness of local opposition also shaped outcomes, so greater openness alone would not have made every operation succeed.
A 39% covert success rate hides which targets were easiest
Success depended on the target. US-backed covert campaigns succeeded 39 percent of the time, compared with 66 percent for overt operations. Covert efforts did better against weak states, democracies, and US allies. Those conditions often gave Washington access to local actors and meant the opposition already had substantial strength. By contrast, attempts to help armed dissidents overthrow governments succeeded only four times in 35 tries.
Tactics mattered too. Nine of 13 US-backed coup attempts succeeded. Parties backed in foreign elections won about three-quarters of the time, although some were already leading and may have won without US help. These figures do not mean that coups or election meddling are reliably effective tools. They show that campaigns needing a small push against an already vulnerable government were more promising than efforts to manufacture a rebellion from scratch.
The book's most important statistical distinction is between winning and causing the win. A campaign can appear successful because a favored party takes office, even if outside assistance made little difference. That is a classic counterfactual problem: the relevant question is what would have happened without the intervention. The author's caution about election campaigns is therefore essential. The figures are still useful for identifying where covert action had favorable conditions, but they should not be mistaken for a recipe that policymakers can reproduce at will.
Replacing leaders rarely removes the pressures that shaped their policies
The successor inherits the dispute. Regime change rests on the hope that a new leader with similar preferences will act in the intervener's interest. But the old government may have resisted because of domestic opposition, national security concerns, or policies essential to its survival. Those pressures do not disappear when the leader changes. The successor may therefore return to the same policies that triggered the conflict.
Dependence has a price. A new leader who follows Washington's wishes may need continued US aid to stay in power. That dependence makes opponents more likely to call the government a foreign puppet, and roughly half of the governments installed through covert US campaigns were later removed violently. Leaders face a familiar bind: comply with the intervener and risk losing local support, or answer local demands and disappoint the sponsor.
This insight reaches beyond foreign policy: changing a decision-maker does little when institutions and incentives continue to produce the same decisions. It also complicates the idea that a shared outlook guarantees cooperation. A leader's preferences are only one influence among domestic coalitions, threats, and administrative limits. The book's account is strongest when it treats imposed leaders as constrained actors rather than simple puppets. One further caution is that dependence can vary over time, so the relationship between aid, legitimacy, and survival is a recurring political contest, not a fixed outcome.
Failed plots can turn covert meddling into durable hostility
Exposure makes failure costly. In more than 70 percent of US Cold War campaigns, Washington was publicly accused of interfering in the target's domestic politics. When an operation failed, the government remained in power with reason to distrust the United States. The book's analysis finds that states targeted by failed campaigns were much more likely to enter a militarized dispute with Washington, and less likely to align with it in international voting.
Targets also paid. States targeted by US covert regime change were more likely to experience civil war and episodes of mass killing afterward. The statistical results are particularly strong for failed operations. The findings do not make every intervention the sole cause of later violence, but they challenge the assumption that failure is harmless when the sponsor can deny involvement.
The striking point is that failure can create a durable information problem. Once a target government suspects foreign sponsorship, even ordinary diplomacy may be read as another attempt at manipulation. That distrust can outlast the original campaign and narrow future options. The book uses matched comparisons to address the possibility that the United States simply targeted countries already prone to conflict. Matching improves the comparison, but it cannot remove every unobserved difference. The evidence supports serious caution, not the claim that covert action alone explains each later war or atrocity.
Diem's 1963 coup shows how a tactical win can deepen entanglement
A leader fell, a crisis widened. In 1963, US officials concluded that Ngo Dinh Diem was too incompetent and politically damaging to lead South Vietnam's fight against communist insurgents. After weeks of debate, Washington supported a coup by South Vietnamese generals. The plot succeeded, but Diem and his brother were killed after their capture, a result Kennedy had not wanted.
The aftermath reversed the hoped-for result. The new leadership lacked Diem's governing structure and domestic base. A series of coups followed, while the Viet Cong took advantage of the instability to intensify its campaign. Rather than stabilizing South Vietnam and making an American withdrawal easier, the operation contributed to a more fragile partner and greater US responsibility. The coup was tactically successful, but that was not the same as achieving the strategic goal.
The case highlights a basic distinction in policy evaluation: removing an obstacle is not equivalent to building a workable replacement. The outcome depended on succession planning, coalition management, and state capacity, not merely on whether the coup reached its immediate objective. That is a powerful lesson for any intervention built around leadership change. The account also resists a simple single-cause explanation of the later war. The coup worsened the political setting, but it operated alongside an existing insurgency and broader US decisions about Vietnam.
Washington backed authoritarian forces in 44 of 64 covert campaigns
Strategic alignment came first. In 44 of 64 covert campaigns, the United States supported authoritarian forces. US policymakers did not consistently favor one kind of regime. They backed dictators, military juntas, or single-party governments when those groups seemed most likely to pursue US interests. They sometimes promoted democracy when officials believed a larger share of the population shared those interests.
Democracy was conditional. Some operations helped parties compete in elections; others sought to replace democratic governments with authoritarian ones. The Dominican Republic provides a clear pattern: Washington supported different arrangements as leaders came and went, but its central aim was to prevent a government hostile to US regional influence. The record challenges the idea that Cold War interventions followed a steady democratic mission. Regime type mattered less than the expected alignment of the people who would hold power.
This evidence separates public ideals from operational priorities without requiring the conclusion that ideals never mattered. Policymakers could sincerely value democracy and still treat it as secondary when they expected an election to produce an unfriendly government. The deeper issue is who gets to define the population's interests: officials often treated alignment with Washington as proof that a regime would serve its citizens. That assumption deserves scrutiny. Elections and institutions are not just instruments for producing compatible leaders; they also give societies a way to contest what compatibility should mean.
Analysis
O'Rourke's central contribution is to make covert regime change visible as a systematic policy choice rather than a collection of notorious episodes. The 64-campaign dataset, built from declassified records and paired with case studies, shows that secrecy was not a side detail. It shaped which operations seemed affordable, which tactics were available, and how officials judged their chances. The book's three-part distinction among rollback, containment, and regional hegemony also explains why anti-communist rhetoric could conceal different strategic aims. Its most compelling causal insight is that leadership is not the same as policy. Governments resist external demands because of security needs, domestic coalitions, and regime survival. A successor inherits these constraints, while dependence on foreign support can undermine the very legitimacy needed to govern. The comparison between the failed attempts against consolidated communist states and successful coups or election campaigns in weaker settings makes this mechanism persuasive. The South Vietnam story gives it human and political texture: removing Diem solved one immediate problem while helping create a more unstable partner.
The evidence warrants care as well as attention. Declassification makes the United States unusually observable, but it means the dataset cannot establish how often other powers use covert regime change or how many successful operations remain undiscovered. Coding campaigns rather than individual acts is sensible, though boundaries between regime change, maintenance, and pressure remain contestable. Statistical matching helps address selection bias, but it cannot recreate the counterfactual or control for every local condition. The numerical associations should therefore be read as cumulative evidence, not mechanical proof of causation.
The book is strongest as a theory of policy incentives and unintended consequences, not as a universal prediction that intervention always fails. Its practical implication is a demanding one: before backing an overthrow, policymakers must show not only why the incumbent is a problem, but how a successor can govern without reproducing the conflict or depending indefinitely on the sponsor. That is a much higher bar than secrecy or a quick victory can satisfy.
People Also Read
Glossary
Regime change
Foreign leadership replacementAn operation intended to replace another state's effective political leadership by substantially changing its ruling elite, administrative apparatus, or political institutions. The target retains its legal sovereignty, unlike in conquest or annexation. The term covers attempts to influence who governs, not every foreign action that affects a state's politics.
Covert regime change
Secretly backed leadership replacementAn operation intended to replace another state's political leadership in which the intervening state does not publicly acknowledge its role. The book includes attempts to assassinate leaders, sponsor coups, influence elections, support revolutions, or assist armed dissidents. The defining feature is concealed state responsibility, not whether the attempt succeeds.
Offensive regime change
Weakening a current rivalA regime change aimed at removing a government regarded as a present military rival or breaking up a rival alliance. During the Cold War, US offensive campaigns were commonly associated with rollback: weakening Soviet power by supporting opposition to Soviet-backed governments. The goal was to reduce a rival's relative power and potentially gain a more cooperative government.
Preventive regime change
Blocking a future threatA regime change aimed at preventing a state from taking actions that could threaten the intervener in the future. Examples include joining a rival alliance, acquiring dangerous military capabilities, or shifting toward an opposing power. In the Cold War, US preventive operations formed part of containment by trying to keep states outside the Soviet camp.
Hegemonic regime change
Preserving regional dominanceA regime change intended to maintain or create a hierarchical relationship between the intervener and a target state. The target need not pose a direct military threat. US campaigns in the Western Hemisphere often reflected a longstanding aim to keep neighboring governments aligned with a US-led regional order and prevent challenges to American predominance.
Plausible deniability
Concealing state responsibilityThe expectation that an intervening government can conceal its role in a covert operation or deny responsibility if its involvement is exposed. Using local political or armed groups as intermediaries is meant to create uncertainty about the sponsor's role. O'Rourke emphasizes that this protection is limited: large operations are difficult to hide, and public accusations were common.
Pseudo-covert operations
Deniable but widely suspected actionsOperations in which a government officially denies involvement even though the target and other observers appear to know or strongly suspect its role. The intervention remains nominally covert because formal acknowledgment would raise political or security costs. The term captures a continuum between fully hidden action and openly acknowledged intervention.
Download PDF
Download EPUB
.epub digital book format is ideal for reading ebooks on phones, tablets, and e-readers.