Tyranny of the Inbox: Managing the US National Security Agenda

Neil N. Snyder
©2025 Neil N. Snyder

ABSTRACT: Presidential management style, foreign policy preferences, and domestic political interests all affect the national security agenda. International crises, however, are particularly likely to garner the attention of the National Security Council. This article analyzes a novel data set of all the issues raised at National Security Council meetings from 1947 to 1993 and finds that contemporaneous crises are very likely to be discussed, but that crisis management attenuates the Council’s attention to noncrisis national security matters. The results suggest presidents focus on crises at the expense of other strategic matters, and they do so when political conditions favor crisis management.

Keywords: national security, presidency, international crises, political science, strategic studies

National security affairs place an enormous burden on US presidents. Waging wars, engaging in diplomacy, and practicing economic statecraft are daunting responsibilities. Managing national security today also means dealing with highly complex issues like quantum computing, artificial intelligence, biodefense, terrorism, cyber, arms control, and nuclear deterrence.

Although many of today’s national security issues are relatively new, challenging national security conditions are not. A litany of global challenges and a barrage of crises plagued the United States at the dawn of the Cold War and may have motivated President Harry S. Truman’s appeals to Congress to reform the national security system. After a fight within the defense community, Congress passed the National Security Act of 1947 and created the modern National Security Council (NSC) system. Today, the NSC is the principal means for presidents to orchestrate national security policy. Presidents cannot and do not personally address every dimension of US national security.¹

In this article, I address the factors that affect the number of issues presidents address and analyze a novel data set of issues raised during presidential NSC meetings from 1947 to 1993, encompassing the Cold War and its immediate aftermath. The historical record offers insights into the national security agenda, conceptualized here as the issues addressed during presidential meetings of the NSC. The NSC takes on a broader agenda when international crises are more prevalent, and presidents are more likely to address crises when their co-partisans have a majority in Congress (for example, during periods of unified government). Presidents also address fewer noncrisis national security issues when the NSC tends to ongoing international crises. These findings account for the unique aspects of each president’s NSC. The NSC’s focus on crises and the consequences of that focus for the breadth of the national security agenda remain persistent features of the NSC system that practitioners must contend with while navigating the complexities of national security policy. This challenge of how to retain strategic breadth amidst crises is the tyranny of the president’s inbox: Presidents focus on crises at the expense of other strategic matters, and they do so when political conditions favor crisis management.

This article addresses a dilemma for national security: tending to crises is often a strategic necessity but may result in less strategic breadth. It also advances a novel explanation for the national security issues that presidents address, considering the confluence of international security conditions, institutional features, and US domestic politics. It tests that explanation with rigorous statistical analysis of a large and comprehensive data set while acknowledging the limits of what the data can reveal.

National security practitioners will find this research significant because the issues addressed by the NSC involve important strategic choices. How, why, and under what conditions issues make it onto the agenda is consequential because NSC deliberations involve presidential decisions about wars, crisis interventions, covert actions, economic coercion, diplomacy, and other national security behaviors of global consequence. Scholars will also find this research significant because it speaks to the broader literature about presidential agenda-setting, executive power, and institutional behavior.

The article begins with an abbreviated case study on the Berlin blockade to motivate thinking about the conditions affecting national security agendas. From there, it builds toward an informal theory of the national security agenda and the NSC and presents three views of the national security agenda. First, the national security agenda is presidential. Second, the agenda is shaped by presidents’ political incentives to attend to international crises. Third, resource constraints affect the breadth of the agenda because of tradeoffs between crisis management and other strategic matters. Then, it introduces a novel data set of all NSC meeting agendas from 1947 to 1993, describes the inference strategy, and presents the results of hypothesis testing. The article concludes with a summary of the findings and a discussion of potential implications.

The Berlin Blockade, Politics, and National Security

At the dawn of the Cold War in the spring of 1948, the United States was entering a new era of global strategic competition with the Soviet Union. At that time, Truman faced reelection and was fighting for his political life: his public approval rating stayed below 40 percent throughout the spring. The post-war economy was flagging, with inflation close to 10 percent and unemployment nearly 4 percent. The Republicans, Truman’s partisan opposition, held the majority in the House and the Senate. In July, the Democratic Party would hold a convention in Philadelphia, Pennsylvania, where Truman expected strong opposition to his nomination as the Democratic presidential candidate in the November elections.²

Truman’s problems ahead of the 1948 convention were not limited to his domestic political misfortunes. A crisis with the Soviet Union was brewing over Berlin, but the NSC system had just been established in the fall of 1947 and was in its infancy. Between January and June 1948, Truman convened the NSC nine times but never held a dedicated NSC meeting about the brewing Soviet situation. The NSC’s inattention to strategic competition with the Soviets that spring was surprising. There had been a communist coup in Czechoslovakia in February 1948. In March, General Lucius D. Clay, the theater military commander, sent an alarmist cable to Washington, DC, warning that the Berlin situation had changed. Clay argued that war with the Soviets “might come with dramatic suddenness.” His prediction was leaked to the press, further exacerbating Truman’s already precarious domestic political situation.³

The Soviets established a complete blockade of Berlin on June 24, 1948, cutting off supplies to US military forces, other Western occupation forces, US citizens, and Berlin’s citizens. Clay, acting without guidance from Washington, initiated operations to sustain Berlin by air, a seminal chapter of the Cold War. Although Clay started the Berlin airlift, his recommendation to the NSC was to resupply Berlin by ground with an armed convoy (a forcible test of the Soviet blockade). Washington saw Clay’s “shoot our way to Berlin” plan as synonymous with starting a war with the Soviets.⁴

Four days after the crisis started, Truman met with the Secretaries of State and Defense for a dedicated session about the situation in Berlin. The Secretaries gave Truman three military options: withdraw, continue the airlift, or prepare for war with the Soviets. Truman chose to stay in Berlin but did not hold a formal NSC meeting on the crisis until July 1. The crisis lasted for 322 days—until May 1949, when the Soviets ended the blockade (though the United States continued the Berlin airlift for four more months, thinking that the Soviets might resume the blockade).⁵

Over the 11-month Berlin crisis, Truman’s NSC met 25 times, including 16 sessions on Berlin. General Omar Bradley, then-Chief of Staff of the Army, later concluded that Truman had failed to give Berlin “his fullest attention” due to the “enormous distraction” of Truman’s domestic political fight. Yet, Truman’s handling of Berlin is hard to fault: he averted war, won reelection (credited by some to his handling of Berlin), and Western allies formed the North Atlantic Treaty Organization shortly before the crisis abated.⁶

In the spring of 1948, Truman faced a hostile Congress, opposition within his political party, unpopularity at home, and the specter of renewed war in Europe. Truman and his advisers gave serious consideration to the use of nuclear weapons if the Soviets escalated further. On the one hand, analysts today might have expected Truman to be particularly attentive to the situation in Berlin because of domestic pressure at home and the national security risks presented by the crisis. On the other, domestic opposition may have shaped whether Truman thought he could act decisively without congressional opposition. Truman was inattentive to Berlin as the situation developed ahead of the Soviet blockade. Yet, Truman’s NSC sustained remarkable focus on the Berlin crisis from the onset until the Soviets lifted the blockade. Truman’s thinking may have been affected by America’s relative military power (then the world’s sole nuclear power), domestic economic challenges, or competing international concerns—such as the Chinese Civil War and the1948 Arab-Israeli War that led to the establishment of the Israeli state, which Truman formally recognized as Israel in May 1948.⁷

The Berlin blockade and Truman’s use of the NSC before and amidst the crisis raise profound questions about the effect of domestic politics and international conditions on when and how presidents engage on national security issues.

What Factors Explain a President’s Agenda?

Since Congress passed the National Security Act of 1947 and created the modern national security system, the NSC has been the president’s forum for national security policy making and implementation, as the one institution designed to “knit it all together.” Although US presidents share agency over national security with Congress, and although our national security capabilities reside with multiple relevant executive branch agencies, the NSC bears the president’s authority and is the focal point for national security matters.⁸

Presidents personally chair formal meetings of the NSC, distinguishing these meetings from principals’ meetings, deputies’ meetings, interagency policy coordinating committee meetings, or others under the Scowcroft system. Presidents have other venues to consult with advisers, such as informal meetings like President Lyndon B. Johnson’s infamous Tuesday lunches, and scholars have presented a mix of quantitative and anecdotal evidence that presidents vary in how frequently they hold NSC meetings. Since 1947, every American president has held formal NSC meetings.⁹

Presidents might address national security matters in other venues, such as during press conferences or State of the Union addresses, but these other means are often platforms to announce policies or actions. In contrast, NSC meetings are deliberative and held in secret, allowing presidents to consider issues in ways that may or may not lead presidents to direct national security actions. The set of issues that presidents deliberate over during NSC meetings, regardless of whether presidents act on those deliberations, is a measure of the president’s national security agenda.¹⁰

Following Kingdon’s theory of the presidential agenda, this conception of the national security agenda is distinctive from the big ideas or intellectual architecture that animates national security policy over long periods, such as grand strategy. The agenda, as a set of issues, is also distinct from individual issue positions or policies that the United States pursues. Whereas issue positions reflect preferences over singular topics, the agenda reflects strategic choices over the set of national security issues that presidents will address during NSC meetings.¹¹

For these reasons, a first view is that the national security agenda is the president’s agenda. Because presidents have strong agency over the NSC and vary in their approaches to national security, national security agendas likely follow the conditions unique to each presidency.

Differences between presidents’ approaches have prompted a rich literature about presidential management styles, including typologies related to presidential experiences, preferences, and delegation patterns, among other dimensions. Presidents have had diverse backgrounds and have varied greatly in their approaches to the NSC. There is evidence that politics shaped the NSC’s formation and evolution, as well as presidents’ ability to exercise control over the national security bureaucracy. Presidents have varied in their approaches to appointments of the NSC staff and how they interact with the NSC staff. This variation extends to presidents’ interactions with the various presidential national security agents, such as the Advisor to the President for National Security Affairs; the Secretary of Defense; senior military leaders such as the Chairman of the Joint Chiefs of Staff; and other cabinet officials, among others to include informal or nongovernmental sources of information and influence on presidents’ thinking. The National Security Advisor may have a particularly significant role in managing the “White House warriors” of the NSC staff. Various factors may also affect the relative influence of different presidential advisers (or the character of presidents’ deliberations over issues with advisers). Nevertheless, the issues addressed at presidential meetings of the NSC are the president’s issues.¹²

This complex of factors suggests that the set of issues addressed at NSC meetings, the strategic choice over the breadth of the national security agenda, is likely to be affected by the unique conditions of each presidency and the peculiarities of each president. Returning to Truman’s NSC during the Berlin blockade as a motivating case, these intuitions would suggest that Truman’s initial aversion to the NSC system and the relative youth of the NSC system in 1948 likely resulted in Truman-unique sets of issues handled at NSC meetings.¹³

Despite these many reasons to believe that the NSC’s function varies by presidency, other factors are likely to affect the breadth of the national security agenda. The uniqueness of each presidency is a condition that we should control for, to make general inferences about what factors influence the breadth of the agenda regardless of who is in office. We should look to other factors that influence the size of the agenda, conditions that practitioners can look to for insight into how the NSC might function as conditions change within or between presidencies.

Consequently, a second view is that the national security agenda likely varies with changing international security conditions. Existing theory suggests that problems, such as international conditions affecting US interests, make the agenda when problems converge with politics and policy making. Moreover, presidents are likely sensitive to the political capital or other resources of the presidency that might be consumed to attend to issues. This logic suggests that the national security agenda is shaped by the nature of issues that arise, presidents’ political incentives and costs to attend to those issues, and how emergent issues interact with existing initiatives and policy planning. Consequently, presidents are likely to take on emergent issues that present opportunities for presidents to take national security actions that might yield political benefits or policy gains toward presidents’ preferences and when the anticipated political costs to take on additional issues are low.¹⁴

Crises, episodes characterized by emergent conditions, uncertainty, high stakes, and threats to national interests, allow presidents to fulfill their national security responsibilities while also taking national security actions—like employing military power—that can have domestic political benefits, such as the well-known “rally ’round the flag effect” (whereby presidents are thought to gain increased public approval for decisive national security actions). While scholars are divided over the conditions under which presidents can expect “rally” effects, research indicates that presidents are more likely to hold NSC meetings when domestic political conditions make attending to international crises politically expedient. Moreover, a sizeable qualitative literature, often relying on insider accounts, attests to the NSC’s focus on international crises. Not only do presidents have incentives to take on crises for political gains, they also have incentives to be attentive to ongoing crises to avoid political losses that can follow disasters like the Bay of Pigs invasion which plagued Kennedy’s relationship with his NSC, or the Iran Hostage Crisis, which beleaguered Carter during his 1980 reelection campaign. Truman likely sought to avoid creating a military disaster over the 1948 Berlin blockade, hence his decision not to take the “shoot our way into Berlin” option preferred by Clay (an option that might have precipitated renewed war in Europe).¹⁵

Given presidents’ incentives to act on crisis opportunities, presidents are likely to address international crises when political or policy gains are possible. However, Congress shares oversight of the national security enterprise, so presidents are likely sensitive to whether Congress will support or oppose the initiatives presidents take through the NSC due to the political costs and resources associated with overcoming congressional opposition (costs that can outweigh the gains presidents otherwise expect from crisis action).

A critical indicator of the president’s anticipated support in Congress is whether the president’s party enjoys a majority in the Senate and the House (unified government), a measure of anticipated political support among elites that matters because of the institutional design of the separation of powers system. When the president’s co-partisans control Congress, the president’s opposition is less likely to be able to mount serious challenges to the president’s agenda. However, under a divided government, serious challenges to the president’s agenda may be more likely, and the president may have to consider exhausting resources to prevent or overcome that opposition (such as outreach to opposition leaders and providing political inducements on other issues, among other options whereby presidents take on resource costs under risk of opposition). Congress can constrain presidents in a variety of ways, from reporting requirements to demanding consultation. Presidents know these constraints and likely consider whether Congress will impose political costs for presidents to pursue their preferred national security agendas. During the Berlin blockade, Truman was likely sensitive to the likelihood of Republican opposition to his national security direction during the first six months of the crisis (until Truman’s party gained a majority in Congress during the November 1948 elections).¹⁶

Together, these ideas build from existing theories of agenda-setting, crisis management, and resources to a theoretical framework for the breadth of the national security agenda. The breadth of the national security agenda is likely to increase with the prevalence of international crises, conditional on the president’s partisan support in Congress. Crises give presidents an opportunity to lead on national security matters and “act presidential,” but leading on national security through the NSC can consume presidents’ resources, including time to generate political support or overcome opposition in Congress. Consequently, presidents’ attentiveness to crises is likely to be tempered by anticipated support or opposition in Congress.

Hypothesis 1 (Crisis Prevalence): The breadth of the national security agenda is likely to increase with the number of ongoing international crises.

Hypothesis 2 (Unified Government): The breadth of the national security agenda is likely to be greater during periods of unified government than under divided government.

Hypothesis 2a (Crisis Prevalence x Unified Government): The effect of crisis prevalence on the breadth of the national security agenda is likely greater during periods of unified government than under divided government.

Increased NSC attention to international crises, as reflected in the breadth of the NSC agenda, may reduce attention to other national security matters. If presidents are careful about expending resources, such as political capital, then taking on emergent issues like international crises may deprioritize other efforts in ways that reduce the rest of the national security agenda.

There are also bureaucratic reasons to think that taking on emergent issues might be costly to the presidency. Staffing the president is intensive. Although NSC staff capacity is robust, emergent national security matters consume NSC staffers’ resources to provide the analysis, options, and policy coordination necessary to support the president’s agenda. Moreover, the prospect of adding new issues on the national security agenda compels the NSC staff (or the president’s other agents at the Departments of State, Defense, and Treasury or within the Intelligence Community) to coordinate positions with other key national security officials and their staffs. The realities of staffing the president for NSC meetings may constrain the NSC’s ability to address emergent issues like crises without impinging on others.

This dilemma is the tyranny of the national security inbox: attending to emergent issues like international crises is likely to reduce effort toward all other national security matters because crises invoke an opportunity cost or tax on the work of the NSC. While the agenda is likely sensitive to the prevalence of national security crises and domestic politics, bureaucratic reality may attenuate the inclusion of non-crises matters.

Moreover, noncrisis matters may not require the same degree of domestic political support as crises. Unlike international crises wherein conditions demand urgent response and political calculus over what Congress will support (or not oppose), noncrisis matters may be more fungible. Presidents and the NSC staff likely have greater leeway to delay or defer noncrisis issues (a consequence of relatively less urgency), and those issues are likely less politically salient than whether to respond to an international threat to US interests under crisis conditions. An example is the contrast between deliberating over a long-running diplomatic negotiation and a pressing crisis: the former might be delayed (or delegated away from the president) without sacrificing potential policy gains or incurring domestic political costs, whereas the NSC may have little perceived ability to delay addressing an ongoing crisis without damaging national security or the president’s domestic standing.

Hypothesis 3 (Crisis Attentiveness): The national security agenda will likely include fewer noncrisis topics when NSC meetings address ongoing crises than when ongoing crises are not addressed.

Data and Analysis

To test these hypotheses, I analyzed the issues addressed at presidential meetings of the NSC from September 18, 1947 (when the modern NSC system was enacted) to January 20, 1993 (at the end of President George H. W. Bush’s administration). The data are the complete record of issues that presidents addressed at NSC meetings for nine presidencies, up to the most recently concluded administration, for which a complete record is available.¹⁷

During this period, presidents held 1,010 NSC meetings. Table 1 presents the number of NSC meetings and NSC topics or issues by presidency. Presidents have varied dramatically in how frequently they convened NSC meetings and how many issues they addressed per meeting. Presidents addressed between 1 and 11 issues per meeting, for a total of 2,738 agenda items over the period, for an average of approximately 2.7 issues per meeting.¹⁸

Although NSC meeting frequency varies between presidencies, table 2 also shows variation in NSC meeting frequency within presidencies, an indication of support for the theoretical framework advanced earlier. Here, the number of NSC meetings is grouped by presidents’ year in office. As a generalization, presidents are likely to hold more NSC meetings early in their tenures than later. Scholars have previously argued that presidents have good reasons to set the agenda early in their administrations, which frees time to wage reelection fights (for first-term presidents) or to cementing legacies (for second-term presidents). Table 2 supports these intuitions, with the caveat that the number of NSC meetings tends to increase early in a president’s second term (years 5-6), and I account for these trends later in the analysis.¹⁹

Table 1. NSC meeting data (1947–93) by council meeting and issue count
(Source: Created by author)

Throughout the period, presidents deliberated over many issues during NSC meetings, from strategic issues related to competition with the Soviet Union, management of ongoing wars, arms control matters, and economic concerns like export control regimes. Figure 1 presents the 25 most frequent NSC meeting topics. Unsurprisingly (given the Cold War), the Union of Soviet Socialist Republics (USSR) was the era’s most frequent NSC agenda issue. International crises feature among the most frequently addressed topics, including the crises over control of Berlin under Truman, Cuba (including the Cuban Missile Crisis and other matters), and Iran (the hostage crisis under Carter).

Table 2. NSC meeting data (1947–93) by meeting count and issue count by presidents’ year in office
(Source: Created by author)

I paired the NSC meeting data with the International Crisis Behavior (ICB) data set to assess variation in the national security agenda related to international crises. The ICB project is an independently assembled set of international crises—episodes characterized by emergent threat, urgency, and uncertainty. For each NSC meeting, I assembled a count of the number of ongoing international crises as a measure of crisis prevalence. During the study period, there were 292 crises, including 47 directly involving the United States. On average, approximately 2.3 international crises were ongoing on NSC meeting days. Figure 2 presents the average number of topics per NSC meeting and the average number of ongoing crises by presidency.²⁰

Figure 2 indicates a strong correlation between the number of ongoing crises and the number of issues presidents addressed at NSC meetings. However, not every crisis made an NSC agenda, and NSC meetings did not cover every contemporaneous crisis, indicating variation in the NSC’s attentiveness to ongoing crises.

Figure 1. Top 25 NSC meeting topics by frequency
(Source: Created by author)

To assess the inclusion of contemporaneous crises on NSC agendas, I compared each meeting’s issue set against contemporaneous ICB crises (by whether meetings fell within the dates of an ICB crisis) by analyzing whether any keywords associated with each crisis of the era were in the issue set for each meeting. This analysis, a dictionary method of text analysis performed by algorithm, enables automated detection of whether ongoing crises made NSC agendas. For example, during Truman’s 1948–49 crisis over the Berlin blockade, I checked whether the terms “Berlin, Germany, Soviet, or USSR” appeared among each meeting’s list of issues. As noted earlier, this algorithmic approach enabled me to identify that Truman discussed Berlin with NSC members 16 times during the 25 NSC meetings held during the Berlin crisis.

Figure 2. Mean number of ongoing crises and issues addressed per NSC meeting presenting 95 percent confidence intervals
(Source: Created by author)

Coding each NSC meeting to measure which NSC agenda items involved contemporaneous crises yields two other measures of the national security agenda: a count of the number of crises addressed at each meeting and the number of noncrisis agenda issues addressed per meeting.

Table 3 presents a cross-tabulation of the NSC meeting data using these measures. Approximately 90 percent of NSC meetings occurred during international crises. When I examined the number of NSC meetings per day for the period of study (the 16,562 days from September 18, 1947, to January 20, 1993, a period that included 14,855 days of crises), there were approximately .06 NSC meetings per day regardless of whether crises were ongoing or not (a rate of approximately 1 NSC meeting every 17 days). NSC meetings were not more likely during crises, despite the ubiquity of crises during the Cold War.²¹

Table 3. NSC meetings and issues by crisis prevalence
(Source: Created by author)

When presidents hold an NSC meeting, they implicitly choose to engage on a set of topics. So, following hypothesis 1 earlier, the question is whether crisis prevalence affects the breadth of the NSC agenda due to presidents’ deliberation over crises. Table 1 shows that presidents address approximately 2.8 topics per meeting during crises, and 2.2 topics per meeting during noncrisis periods. This evidence is initial empirical support for hypothesis 1.

Table 4 further disaggregates the 901 crisis-period NSC meetings into two types: when ongoing crises made the agenda and when they did not. While only one-third of crisis-period NSC meetings addressed contemporaneous crises (236 of 901 meetings), presidents averaged 3.4 crisis-related issues or topics per NSC meeting when crises made the agenda and 2.6 crisis-related topics when crises were not on the table. This result might initially appear unsurprising, but there were between two and three ongoing crises on most days of the Cold War (an average of 2.4 per day), suggesting that much less than half of ongoing crises make NSC agendas. This inattention to crises is surprising, given the generally heightened risk perceptions of American leaders during the Cold War. Later analysis explores this relationship further.

Table 4. Crisis-period NSC meetings by type and number of NSC issues by type
(Source: Created by author)

The number of noncrisis-related issues is also telling. During noncrisis periods, presidents averaged 2.2 issues per meeting. During crises, but when ongoing crises did not make the agenda, presidents averaged 2.6 noncrisis topics per meeting, suggesting that presidents are generally more sensitive to national security matters during crises (even when they do not address crises during NSC meetings). Finally, during meetings that addressed ongoing crises, presidents averaged 2.2 noncrisis topics: significantly fewer noncrisis topics are addressed when presidents attend to crises, as compared to crisis-period meetings that do not address ongoing crises. This difference is small on the scale of a single NSC meeting but becomes a non-trivial difference when the number of noncrisis topics is extrapolated over many NSC meetings. Over time, crisis myopia significantly reduces the number of noncrisis topics the NSC addresses. Although this summary data does not control for other factors, this observed crisis myopia provides initial empirical support for hypothesis 3, the intuition that crisis attention during NSC meetings may attenuate the noncrisis part of the president’s national security agenda.²²

A final set of contrasts for this NSC meeting and issue data concerns presidents’ support in Congress. Table 5 contrasts the number of NSC meetings during unified government with the number under divided government. Most Cold War NSC meetings occurred during divided government. The data suggests, however, that NSC meetings addressed more topics per meeting under unified government. This correlation between unified government and increased NSC agenda breadth provides initial empirical support for hypothesis 2, the claim that presidents’ attention to national security issues may increase when Congressional opposition is less likely or, conversely, when support in Congress is more likely due to co-partisan control of Congress).²³

Table 5. NSC meetings and issues by government type
(Source: Created by author)

This description of the data yields several insights. First, the NSC addressed more topics during crises than noncrisis periods, regardless of whether ongoing crises make the agenda, suggesting that crisis conditions increase the NSC’s overall sensitivity to national security issues. Second, the NSC addresses fewer noncrisis topics when crises make the agenda (compared to the relevant counterfactual condition, a crisis-period meeting that does not address a contemporaneous crisis). Although this data is descriptive, it provides initial empirical support to the hypothesis for the tyranny of the inbox (that crisis-attentiveness attenuates the inclusion of noncrisis topics on the agenda).

Statistical Results

The number of issues per NSC meeting is a compound outcome, and various factors may affect the issue set placed on the national security agenda. To address this concern that many factors other than crisis conditions might affect NSC agendas, I estimated the effect of key variables (crisis prevalence and unified government) on the number of crisis and noncrisis issues per meeting while controlling for the unique aspects of each presidency and each president’s time in office (fixed effects or separate intercepts, which in theory account for all time-invariant conditions common to each presidency and each president’s year of office).²⁴

I used a series of ordinary least squares regression models that test the effect of crisis prevalence and government type on the number of issues presidents address at NSC meetings. These results confirm what the descriptive data already shows: NSC agenda breadth increases with crisis prevalence and the NSC addresses fewer noncrisis issues when the NSC is crisis attentive. For nontechnical readers, I summarize the findings at the beginning of the next section.

To account for other factors that might affect the issues raised at NSC meetings, I assembled a battery of statistical controls (tests for whether the estimated effects of unified government or crisis-prevalence are robust to possible other sources of variation in NSC meeting issue counts). These variables relate to three broad categories of influence on national security behavior—domestic politics, political economy, and international security. Statistical controls included presidents’ approval ratings, reelection periods (defined here from the major party primaries started until election day), economic indicators (unemployment and the consumer price index, a measure of inflation), and periods of ongoing war (Korea, Vietnam, and the Gulf War were the three major wars of the period involving the United States). Additionally, because existing studies suggest that presidential notational security behavior may vary with the US relative power, I include hegemony, a measure of the US share of global power, an annual index on a percentage scale from the composite index of national capability found in the National Military Capabilities data set, version 6.0.²⁵

Table 6 presents the results of statistical tests for whether crisis prevalence and periods of unified government affect the number of issues that make the US national security agenda, as measured by the number of issues presidents address during NSC meetings.²⁶

Models 1 and 2 test the effect of various conditions on the number of crisis-related issues raised during NSC meetings. Here, we find that crisis prevalence and unified government are associated with a statistically significant increase in the number of crisis-related agenda items, evidence in support of hypotheses 1 and 2. Model 2, for example, suggests each additional ongoing crisis is associated with an approximate 10 percent increase in crisis-related topics on NSC agendas. Unified government is associated with approximately 0.4 more topics per meeting than divided government. While the practical magnitude of this effect at the unit of a single NSC meeting is small (because NSC meetings average two to three issues per meeting), over time, this effect aggregates to the large differences observed earlier in table 4. Models 1 and 2 also indicate the number of crisis-related issues at NSC meetings increases with periods of war (when national security is already at risk), increases when US relative power is low (as indicated by the negative coefficient on “Hegemon”), and increases with inflation (as per the coefficient on CPI). These results comport with a traditional view of the NSC as a holistic forum for presidential national security management.²⁷

Models 3 and 4 address the conditions affecting the number of noncrisis issues per NSC meeting. Because the outcome variable here is the number of noncrisis issues per meeting and, given that some meetings involve both crisis-related topics and noncrisis topics, these models allow introducing a new variable, crisis attention, which captures whether noncrisis topics shared agenda space with crisis matters during specific NSC meetings. In model 4, crisis attention during NSC meetings is associated with .6 fewer noncrisis topics per meeting, evidence that supports hypothesis 3. That effect is significant when extrapolated across the 236 NSC meetings of the period that involved crisis attention, resulting in approximately 141 fewer total noncrisis agenda items than might otherwise have been taken to the president (had crisis attention not been a negative influence on noncrisis topics’ inclusion at NSC meetings).

Noncrisis issue count is also uncorrelated with other variables related to security risk (wartime, hegemony) consistent with a view of noncrisis topics as qualitatively different than crisis issues. Importantly, model 4 shows that the count of noncrisis topics on NSC agendas is not correlated with political variables (unified government or approval rating) at conventional levels of statistical significance, suggesting that politics does not play into when presidents address noncrisis matters. The insignificant effect of political conditions on noncrisis NSC topics strongly contrasts model 2, wherein congressional competition affects the NSC’s attention to crises. Paradoxically, the topics involving the greatest risk and urgency (crises) are conditional on politics in Congress, whereas noncrisis matters are not.

Table 6. Ordinary least squares (OLS) regression models of NSC topic counts
(Source: Created by author)

Altogether, the statistical results presented in table 6 support the earlier hypotheses, particularly regarding NSC issues by type. I find that crisis prevalence and unified government likely increase the number of crisis issues addressed during NSC meetings, but also that attending to crises has the predictable effect of reducing the attention paid to noncrisis issues.

These results have limitations. To address readers’ potential concerns with the main results, I performed multiple tests to assess the robustness of the results. I have omitted those results for brevity but will describe some of the tests briefly to reassure readers.²⁸

My robustness checks included controlling for the tenure of each National Security Advisor during the period, on the intuition that presidents’ National Security Advisors strongly influence NSC operations. Second, I estimated alternative models on data limited to US, NATO, or Soviet-involved crises on the intuition that only crises were salient during the Cold War and might be more likely to drive NSC agendas. Third, I estimated models while accounting for protracted or recurring crises of the era (such as the Taiwan Straits or the various Arab-Israeli conflicts) to assess whether agenda breadth was equally affected by truly emergent crises and by recurrent or known flashpoints. Fourth, I estimated models using Poisson regression (a model specifically designed for count data), an alternative to the linear models used in the main results. Each of these tests, as well as others not detailed here for brevity, affirmed the main results in table 6.

Discussion

This article yields two main findings. First, the summary data and modeling results suggest that crisis prevalence and periods of unified government are likely to result in increased crisis attentiveness during presidential NSC meetings. Second, noncrisis issues are significantly less likely to make NSC agendas when presidents address ongoing crises.

Although these findings might seem obvious to national security practitioners, they provide insight into a consequential feature of US national security decision making that is prevalent across the declassified history of the NSC: focusing on crises constrains the president’s strategic breadth. This consequence should guide how practitioners approach strategic issues during future periods of crisis. Issues discussed at the NSC can lead to presidentially directed actions to employ the US instruments of national power for strategic effect or, when issues are deferred, inaction with arguably equal strategic implications.

These results have limitations. First, the focus is on the US NSC. As an institution, the NSC is unique to the American political system, and the results likely would not hold for other countries (even for other mature democracies with established national security advisory systems). Moreover, the data is from the Cold War, and it has been more than 30 years since it ended. Given China’s rise and Russia’s revanchist aggression in Ukraine, today’s environment has unique features that depart from the conditions of the Cold War. Future research may benefit from extending this research into the post–Cold War period, when additional records become available from the NSC, to test whether the results hold for later periods.²⁹

Second, there are good reasons to be concerned with endogeneity between NSC agendas at one time, international conditions that follow from NSC-directed actions, and the NSC work that follows. This inter-relatedness is relevant when considering that the NSC’s work could affect the duration of crises (NSC-directed actions like an intervention or NSC inaction could reduce, perpetuate, or otherwise change conditions affecting how long a crisis lasts). While the modeling approach takes reasonable steps to mitigate concern for bias, the models assume each observation (in this case, each NSC meeting) is independent. Future research may benefit scholars by employing qualitative methods or advanced text research to assess the continuities between NSC agenda topics and how NSC actions affect crisis duration or outcomes. Moreover, the analysis excludes how presidents might engage with national security matters in non-NSC fora, such as during informal meetings with cabinet members or advisers. Future work may benefit from using case studies to compare presidents’ informal (non-NSC) deliberations with their NSC discussions during crises. If scholars and practitioners could identify the issues that are only addressed informally or only addressed in NSC meetings, we may learn more about how presidents operate.

Third, this research addresses NSC issues during crises. There are good questions of whether NSC attentiveness to specific strategic issues might help avoid crises from occurring (precrisis measures to prevent or mitigate) as well as good questions of whether there is a hangover effect from NSC crisis attentiveness (post-crisis effects on the NSC’s work). This research does not address those questions, though these are promising areas for future research.

Finally, future research may benefit from studying the issues addressed in NSC meetings and the national security actions that an administration takes. Practitioners and scholars would benefit from understanding whether crisis severity and NSC crisis attentiveness affect the kinds of national security actions that presidents take.

The results have significant implications for national security scholars and practitioners. One implication is that because presidents are less likely to attend to international crises during a divided government, scholars and practitioners may benefit from accounting for a diminished US national security agenda that during periods of divided government. Divisive domestic politics undermine US foreign policy decision making, conditions likely exacerbated when presidents defer from addressing specific issues. Specifically, there is a vital question of whether periods of diminished national security focus due to divided government (or just contentious politics) might contribute to or aggregate into a progressive strategic decline for the United States in ways that complicate its standing in the international order. Moreover, inattention to a broad range of strategic matters may make future crises more likely. As Kissinger noted, “in high office competing pressures tempt one to believe that an issue deferred is a problem avoided; more often it is a crisis invited.”³⁰

A second implication is that practitioners will likely benefit from appreciating how the tyranny of the inbox can attenuate the NSC’s attention on noncrisis matters—a consequence of domestic institutions and international conditions. Considering known challenges with US national security planning and national security elites’ suggestion that presidents must “protect time on the schedule” for regular NSC meetings, practitioners would likely benefit from incorporating measures to structure NSC deliberations in ways that enable sustained attention to strategic matters that have yet to become crises.³¹

Finally, scholars and practitioners may benefit from applying the insights from this Cold War NSC data to the contemporary era of great-power competition. Crises were pervasive during the Cold War competition with the USSR, and when presidents addressed those crises through the NSC, attention to noncrisis national security matters suffered. Politics and competing national security commitments can affect the US ability to pursue national security priorities amidst emergent challenges. Organizing national security processes to account for this dilemma may help the US national security community “walk and chew gum,” as articulated by former Secretary of Defense Lloyd Austin in his description of the US ability to simultaneously manage competition with China and reassure Europe amidst Putin’s aggression in Ukraine. The United States will need to find ways to deal with ongoing crises like Ukraine, Gaza, or others without sacrificing the overall management of strategic competition.³²

Neil N. Snyder
Colonel Neil N. Snyder, PhD, is an assistant professor in the Department of National Security Strategy at the US Army War College. He holds a PhD in political science from Stanford University as a fellow of the US Army Strategic Plans and Policy Program, a master in military art and science degree from the US Army Command and General Staff College, a master of science degree in defense analysis from the Naval Postgraduate School, and a bachelor of science in engineering degree in biomedical engineering and economics from Duke University.

Disclaimer: Articles, reviews and replies, review essays, and book reviews published in Parameters are unofficial expressions of opinion. The views and opinions expressed in Parameters are those of the authors and are not necessarily those of the Department of Defense, the Department of the Army, the US Army War College, or any other agency of the US government. The appearance of external hyperlinks does not constitute endorsement by the Department of Defense of the linked websites or the information, products, or services contained therein. The Department of Defense does not exercise any editorial, security, or other control over the information you may find at these locations.

Endnotes

1. Amy Zegart, Flawed by Design: The Evolution of the CIA, JCS, and NSC (Stanford University Press, 1999), 61, 67. Return to text.
2. “Presidential Approval Rating Data Set,” 2021, Roper Center for Public Opinion Research, https://presidential.roper.center/; “Bureau of Labor and Statistics,” 2022, https://data.bls.gov; United States Senate “Party Division,” 2021, https://www.senate.gov/history/partydiv.htm; “Party Divisions of the United States House of Representatives,” 2021, https://history.house.gov/Institution/Party-Divisions/Party-Divisions/. Return to text.
3. Omar N. Bradley and Clay Blair, A General’s Life: An Autobiography by General of the Army Omar N. Bradley (Simon & Schuster, 1983), 477; and Peter Grose, “The Boss of Occupied Germany: General Lucius D. Clay,” Foreign Affairs 77, no. 4 (1998): 179–85. Return to text.
4. Jean Edward Smith, Lucius D. Clay: An American Life (Henry Holt and Co., 1990), 499; “A Report to the National Security Council by the Secretary of Defense on U.S. Military Course of Action with Respect to the Situation in Berlin” (National Security Council, July 28, 1948); and Bradley and Blair, A General’s Life, 479. Return to text.
5. Bradley and Blair, A General’s Life, 478; and Wilson D. Miscamble, “Harry S. Truman, the Berlin Blockade, and the 1948 Election,” Presidential Studies Quarterly 10, no. 3 (1980): 306–16. Return to text.
6. Robert T. Davis II, ed., U.S. Foreign Policy and National Security: Chronology and Index for the 20th Century, vol. 1 (Praeger, 2010), 275–84; Bradley and Blair, A General’s Life, 479; and Miscamble argues that Truman did not consult with political advisers over his Berlin blockade decisions and that there was no evidence that Truman deliberately acted to curry favor with voters, though his successful handling of the crisis neutered partisan criticism of his foreign policy acumen. Miscamble, “Harry S. Truman,” 311–12. Return to text.
7. Walter Isaacson and Evan Thomas, The Wise Men: Six Friends and the World They Made (Simon & Schuster, 1986), 460. Return to text.
8. Paul D. Miller, “The Contemporary Presidency: Organizing the National Security Council: I Like Ike’s,” Presidential Studies Quarterly 43, no. 3 (2013): 593; and R. Gordon Hoxie, “The National Security Council,” Presidential Studies Quarterly 12, no. 1 (1982). Return to text.
9. Miller, “The Contemporary Presidency” 592–95; John W. Rollins, The National Security Council: Background and Issues for Congress, Congressional Research Service (CRS) Report 44828 (CRS, October 19, 2022); and David Auerswald, “The Evolution of the NSC Process,” in The National Security Enterprise: Navigating the Labyrinth, ed. Roger Z. George and Harvey Rishikof (Georgetown University Press, 2017), 31–54; Alexander L. George, Presidential Decisionmaking in Foreign Policy: The Effective Use of Information and Advice (Avalon Publishing, 1980), 145–58; Morton H. Halperin, Bureaucratic Politics and Foreign Policy (Brookings Institution Press, 1974), 105, 109; Peter W. Rodman, Presidential Command: Power, Leadership, and the Making of Foreign Policy from Richard Nixon to George W. Bush (Knopf Doubleday Publishing Group, 2009); and David M. Barrett, “Doing ‘Tuesday Lunch’ at Lyndon Johnson’s White House: New Archival Evidence on Vietnam Decisionmaking,” PS: Political Science and Politics 24, no. 4 (1991): 676–79. Return to text.
10. National Security Council secrecy is likely a good reason why there is limited existing empirical scholarship on the NSC. Andrew Rudalevige, The New Imperial Presidency: Renewing Presidential Power After Watergate (University of Michigan Press, 2005), 174–75. However, as will be evident in the empirical section, NSC records do become available after US governmental declassification processes, which generally restrict availability of classified records for 25 years; “A Rule by the Federal Emergency Management Agency on 07/30/2014,” Exec. Order 13526 (Classified National Security Information), Federal Register 72, no. 2 (December 29, 2009): 707–31. As secret governmental documents are released, large volumes of data become available for scholars to analyze. Matthew Connelly, The Declassification Engine: What History Reveals About America’s Top Secrets (Pantheon Books, 2023). Return to text.
11. John W. Kingdon, Agendas, Alternatives, and Public Policies, updated 2nd ed. (Longman, 2011), 3; Hal Brands, What Good Is Strategy?: Power and Purpose in American Statecraft from Harry S. Truman to George W. Bush (Cornell University Press, 2014); Richard K. Betts, “Is Strategy an Illusion?,” International Security 25, no. 2 (2000): 5–50; and Nina Silove, “Beyond the Buzzword: The Three Meanings of ‘Grand Strategy,’ ” Security Studies 27, no. 1 (2018): 27–57. Return to text.
12. Richard Tanner Johnson, Managing the White House: An Intimate Study of the Presidency (Harper & Row, 1974); Richard E. Neustadt, Presidential Power and the Modern Presidents (Macmillan, 1960); Thomas Preston, The President and His Inner Circle: Leadership Style and the Advisory Process in Foreign Affairs (Columbia University Press, 2001); Michael C. Horowitz and Allan C. Stam, “How Prior Military Experience Influences the Future Militarized Behavior of Leaders,” International Organization 68, no. 3 (2014): 527–59; Michael C. Horowitz et al., Why Leaders Fight (Cambridge University Press, 2015); Elizabeth N. Saunders, Leaders at War: How Presidents Shape Military Interventions (Cornell University Press, 2011); George, Presidential Decisionmaking; Halperin, Bureaucratic Politics; Hoxie, “Security Council”; Zegart, Flawed by Design; Auerswald, “NSC Process”; Charles P. Ries, “How Did the National Security System Evolve?,” in Improving Decisionmaking in a Turbulent World (RAND Corporation, 2016), 11–22; Rodman, Presidential Command; John P. Burke, “The National Security Advisor and Staff: Transition Challenges,” Presidential Studies Quarterly 39, no. 2 (2009): 283–321; Robert Cutler, “The Development of the National Security Council,” Foreign Affairs 34, no. 3 (1956); John Gans, White House Warriors: How the National Security Council Transformed the American Way of War (Norton, 2019); Tyler Jost et al., “Advisers and Aggregation in Foreign Policy Decision Making,” International Organization 78, no. 1 (2024): 1–37; Saunders, Leaders at War; Elizabeth N. Saunders, “War and the Inner Circle: Democratic Elites and the Politics of Using Force,” Security Studies 24, no. 3 (2015): 466–501; Elizabeth N. Saunders, “Leaders, Advisers, and the Political Origins of Elite Support for War,” Journal of Conflict Resolution 62, no. 10 (2018): 2118–49; Elizabeth N. Saunders, “No Substitute for Experience: Presidents, Advisers, and Information in Group Decision Making,” International Organization 71, no. S1 (2017): S219–47; George, Presidential Decisionmaking. Return to text.
13. Zegart, Flawed by Design. Return to text.
14. Kingdon, Agendas, Alternatives, and Public Policies; and Paul Charles Light, The President’s Agenda: Domestic Policy Choice from Kennedy to Clinton, 3rd ed. (John Hopkins University Press, 1999). Return to text.
15. Michael A. Genovese, “Presidential Leadership and Crisis Management,” Presidential Studies Quarterly 16, no. 2 (1986): 301; George, Presidential Decisionmaking; William D. Baker and John R. Oneal, “Patriotism or Opinion Leadership? The Nature and Origins of the ‘Rally ’Round the Flag’ Effect,” Journal of Conflict Resolution 45, no. 5 (2001): 661–87; Matthew A. Baum, “The Constituent Foundations of the Rally-Round-the-Flag Phenomenon,” International Studies Quarterly 46, no. 2 (2002): 263–98; Terrence L. Chapman and Dan Reiter, “The United Nations Security Council and the Rally ’Round the Flag Effect,” Journal of Conflict Resolution 48, no. 6 (2004): 886–909; John. E. Mueller, War, Presidents, and Public Opinion (John Wiley & Sons, 1973); Shoon Murray, “The ‘Rally-’Round- the-Flag’ Phenomenon and the Diversionary Use of Force,” in Oxford Research Encyclopedia of Politics (Oxford University Press, 2017), https://oxfordre.com/politics/view/10.1093/acrefore/9780190228637.001.0001/acrefore-9780190228637-e-518; Bruce M. Russett, Controlling the Sword: The Democratic Governance of National Security (Harvard University Press, 1990); Neil Snyder, “National Security in Presidential Time: The Politics of the National Security Council,” Presidential Studies Quarterly 53, no. 4 (2023): 517–42, https://doi.org/10.1111/psq.12849; Neil Snyder, “Presidential Attentiveness to International Crises,” Presidential Studies Quarterly, 2024, https://doi.org/10.1111/psq.12875; Michael Bohn, Presidents in Crisis: Tough Decisions Inside the White House from Truman to Obama (Arcade Publishing, 2015); Gans, White House Warriors; Ralph Gordon Hoxie, Command Decision and the Presidency: A Study in National Security Policy and Organization (Reader’s Digest Press, 1977); John Prados, Keepers of the Keys: A History of the National Security Council from Truman to Bush (William Morrow & Company, 1991); William G. Howell and Jon C. Pevehouse, While Dangers Gather: Congressional Checks on Presidential Powers (Princeton University Press, 2007), 22; and Stephen Skowronek, The Politics Presidents Make: Leadership from John Adams to Bill Clinton (Harvard University Press, 1993), 401. Return to text.
16. Howell and Pevehouse, While Dangers Gather; Douglas L. Kriner, “Congress, Public Opinion, and an Informal Constraint on the Commander-in-Chief,” The British Journal of Politics and International Relations 20, no. 1 (2018): 52–68; James Nathan, “Salvaging the War Powers Resolution,” Presidential Studies Quarterly 23, no. 2 (1993): 235–68; William G. Howell, Power Without Persuasion: The Politics of Direct Presidential Action (Princeton University Press, 2003), 105; William G. Howell and Jon C. Pevehouse, “Presidents, Congress, and the Use of Force,” International Organization 59, no. 1 (2005): 209–32; and Howell and Pevehouse, While Dangers Gather. Return to text.
17. I assembled NSC records from Robert T. Davis’s national security anthology Davis, U.S. Foreign Policy and National Security: Chronology and Index for the 20th Century, 2010, vol. 2 (Praeger, 2010), after checking against other public sources for accuracy, including “National Security Archive,” 2021, https://nsarchive.gwu.edu/; and “National Security Council Historical List of NSC Meetings with Agenda Topics,” 2021, https://clintonwhitehouse4.archives.gov/WH/EOP/NSC/html/historical/Meetings.html. I constructed a data set from these NSC meeting records to enable statistical analysis. The data assembled for this article, and associated scripts or coding for statistical analysis, is available on request from the author. Davis’s records provide post-meeting accounts of topics addressed at NSC meetings, topics that are distilled from primary source documents, including archival copies of NSC meeting minutes at presidential libraries. The data is the set of topics (discrete areas of discussion during an NSC meeting), not the complete NSC meeting minutes, which are not uniformly available across the period of study as of this writing. Because source records come from declassified sources such as government-internal records that were previously classified or restricted from public access for reasons of US national security, there is little reason to believe topics have been omitted or excluded from the declassified record. By the US declassification guidelines, records are declassified after 25 years unless specific exceptions are applied, according to Classified National Security Information, Executive Order 13526. Analysis of completed/concluded presidencies with declassified sources precludes analysis beyond the George H. W. Bush administration until further declassified records are made available to researchers. Return to text.
18. The number of issues or topics addressed per NSC meeting is a measure of the breadth the NSC’s work. An alternative might be to measure the duration (in hours and minutes) of NSC meetings, or the time spent per topic. While issue counts measure whether the NSC addressed certain topics, the duration of meetings (or duration on topics) measures or characterizes how the issue was addressed. That is because duration likely reflects other factors like the amount of information the NSC had on a topic, the complexity of that topic, or perhaps the degree of NSC principals’ consensus on a topic. For example, we might think a topic got less attention if a meeting on that topic was brief, but instead a concise meeting might follow because there were limited strategic options or strong consensus on the topic. To prevent the measurement from being distorted by these other factors, the analysis here focuses on a simple measure of number and type of issues raised per NSC meeting. Return to text.
19. Office years are calculated as sequential 365-day periods from inauguration or assumption of office, to allow fair comparison of tenures for presidents who assumed office offset of the normal electoral cycle (Truman, Johnson, Gerald Ford) due to their predecessors’ deaths (Theodore Roosevelt, John F. Kennedy) or resignations (Richard Nixon); Burke, “Transition Challenges,” 314; Light, President’s Agenda, 38–46; Philip A. Odeen, “Organizing for National Security,” International Security 5, no. 1 (1980): 114; and Russett, Controlling the Sword, 47. Return to text.
20. Michael Brecher et al., “International Crisis Behavior Data Codebook, Version 14,” 2021, https://duke.box.com/s/nttm3e1yfyv4kbacntj5g2pqi5by2zyy; and Michael Brecher and Jonathan Wilkenfeld, A Study of Crisis (University of Michigan Press, 2000). Return to text.
21. While prior statistical results suggested that NSC meeting frequency is positively correlated with the number of ongoing military disputes (particularly acute episodes involving the use of military force at onset), NSC meeting frequency is not correlated with ICB crisis frequency at any conventional level of statistical significance. Snyder, “National Security in Presidential Time.” Return to text.
22. A two-tailed t-test for the difference in means, between noncrisis topic counts by meeting type, suggests a statistically significant difference (at p= 0.008). Return to text.
23. “Party Division of the United States Senate”; “Party Division of the United States House of Representatives”; Prior research suggests NSC meeting frequency is not affected by periods of unified or divided government with statistical significance (the positive correlation between NSC agenda issues and unified government is not likely to be biased by meeting frequency. Snyder, “National Security in Presidential Time.” Return to text.
24. Including fixed effects for each presidency and each office year provides a panel-like estimator, controlling for conditions across each presidency and over time within each presidency. Additional models were estimated with only fixed effects for each presidency. Those results, excluded for brevity, affirm the main results presented later in table 6. Return to text.
25. “Presidential Approval Rating Data Set.”; “Bureau of Labor and Statistics”; Coding for wartime periods is as follows. The Korean War was from onset in June 1950 until armistice in late July 1953. The Vietnam War was from the Tonkin Resolution in 1964 until US troop withdrawals were completed in 1973. The Gulf War started with Iraq’s invasion of Kuwait in August 1990, which prompted Operation Desert Shield, through cessation of US-led combat operations at the end of February 1991. Barbara Salazar Torreon, U.S. Periods of War and Dates of Recent Conflicts, CRS Report RS21405 (CRS, June 5, 2020); J. David Singer, “Reconstructing the Correlates of War Dataset on Material Capabilities of States, 1816–1985,” International Interactions 14, no. 2 (1987): 115–32; J. David Singer et al., “Capability Distribution, Uncertainty, and Major Power War, 1820–1965,” in Peace, War, and Numbers, ed. Bruce M. Russett (Sage, 1972), 19–48. Return to text.
26. All modeling was completed using the R statistical computing language and the Fast Fixed Effects package for R. Laurent Berge, “Efficient Estimation of Maximum Likelihood Models with Multiple Fixed-Effects: The R Package FENmlm,” CREA Discussion Papers, 2018; and “R: A Language and Environment for Statistical Computing” (2023), http://www.R-project.org. Return to text.
27. In addition to the model 2 results, additional models were estimated with the interaction of the ongoing crisis count and unified government, to test formally whether unified government moderates the effect of crisis prevalence on the agenda count of crisis issues. Those results, omitted for simplicity of presentation and interpretation here, show the main effect of crisis count and the interaction term are statistically significant (p<0.05), consistent with the view that government type (unified or divided) moderates presidents’ attentiveness to ongoing crises at NSC meetings. The significance of the interaction effect (of crisis prevalence by unified government) is support for hypothesis 2a. Return to text.
28. Results for these robustness checks, along with other supplementary results, are available on request. Return to text.
29. Bilahari Kausikan, “Navigating the New Age of Great-Power Competition: Statecraft in the Shadow of the U.S.-Chinese Rivalry” Foreign Affairs 82, no. 2 (April, 2023), https://www.foreignaffairs.com/united-states/china-great-power-competition-russia-guide. Return to text.
30. Kenneth Schultz, “Perils of Polarization for U.S. Foreign Policy,” Washington Quarterly 40, no. 4 (2017): 7–28; and Henry Kissinger, The White House Years (Simon & Schuster, 1979), https://content.time.com/time/subscriber/article/0,33009,916938-13,00.html. Return to text.
31. Aaron L. Friedberg, “Strengthening U.S. Strategic Planning,” Washington Quarterly 31, no. 1 (2007): 47–60; Robert Jervise, “U.S. Grand Strategy: Mission Impossible,” Naval War College Review 51, no. 3 (1998): 22–36; Amy Zegart, “Why the Best Is Not Yet to Come in Policy Planning,” in Avoiding Trivia: The Role of Strategic Planning in American Foreign Policy, ed. Daniel W. Drezner (Brookings Institution Press, 2009), 113–24; and Michele A Flournoy, Nine Lessons for Navigating National Security (Center for New American Security, 2016), http://www.jstor.com/stable/resrep06210. Return to text.
32. Jim Garamone, “Austin: How the U.S. Walks, Chews Gum at the Same Time,” DoD News, June 13, 2022, https://www.defense.gov/News/News-Stories/Article/Article/3061381/austin-how-the-us-walks-chews-gum-at-the-same-time/. Return to text.