Long-term Thinking in U.S. National Security

The case of the U.S. Coast Guard shows how strategic foresight can inform strategy, which can in turn influence operations. But foresight, strategy, and operations are distinct activities, whose symbiotic relationship is by no means assured. Organizations may conduct foresight activities but fail to derive strategy from them, and even carefully articulated strategies may have little connection to operations. Based on interviews with current and former U.S. officials, as well as a review of primary and secondary documents, this section disaggregates these activities by examining the American national security establishment, the collection of organizations responsible for protecting the United States at home and advancing its interests abroad. It shows that even a plethora of “strategy” does not necessarily yield policies that connect visions of the future to actions in the present.

The national security establishment stands out in the federal government not only because it is huge—by one estimate the United States spends $1.25 trillion annually on national security¹—but also because, to function properly, it must make an unusual number of high-stakes, long-term decisions without a clear view of what the long term looks like. Cultivating diplomatic arrangements; assuring basing and overflight rights; developing, testing, and fielding military forces; and training and developing a federal workforce that has the appropriate skills are all long-term propositions demanding long-term plans.² As a result, national security practitioners, especially Pentagon officials, are often forward-looking. But long-term plans are only as good as the accuracy of long-term predictions, and as management scholar Henry Mintzberg wrote, it is a fallacy to believe that the world will “hold still while a plan is being developed and then stay on the predicted course while that plan is being implemented.”³

The geopolitical far-future is particularly uncertain because of the complexity of the international system, and foreign policy experts have a lousy predictive record.⁴ As Robert Gates said in 2011, while he was serving as President Obama’s secretary of defense, “When it comes to predicting the nature and location of our next military engagements, since Vietnam, our record has been perfect. We have never once gotten it right.”⁵ The U.S. intelligence community has tried to predict the future since Sherman Kent led the Office of National Estimates in the CIA’s early years.⁶ Yet, despite the allocation of significant time, money, and effort, the intelligence community has often failed to anticipate the near-term future, let alone the long-term future. In 1973, Gates—then a young intelligence analyst—wrote for the agency’s in-house journal: “We failed to anticipate the construction of the Berlin Wall, the ouster of Khrushchev, the timing of the invasion of Czechoslovakia, and other events of importance.”⁷ Later surprises would include the Iranian revolution, the fall of the Berlin Wall, and the collapse of the Soviet Union.

(Ironically, when he served as deputy director of the CIA, Gates himself erred on one of the most important judgments the agency faced: whether Mikhail Gorbachev was a reformer.)⁸

The point is not that the foreign policy, defense, or intelligence communities are inept. The point is that, as discussed above, quantity of thought does not necessarily equal quality of thought, let alone accuracy of anticipation. After all, one can think about the future, but think about it badly. In his study of the highly structured “strategic planning” programs that guided large agencies like the Pentagon in the postwar years, Mintzberg asked, “Does formal recognition of the future, let alone formalizing how it is dealt with, necessarily mean the future is properly taken into account?”⁹ It does not. Richard Fontaine, the CEO of the Center for a New American Security, gave one conspicuous example, pointing out that thinking about the long term did not yield foreknowledge in the run-up to the Iraq war:

Key movers in the Bush administration did think long-term, but about the wrong things: the potential for a democratic Iraq to spread its political system to other Middle East autocracies, making the region ultimately more democratic, less ridden by terrorists, and better for both the U.S. and people in the region. Their failure was in not thinking about and planning for the other, less attractive scenarios, which were much likelier.¹⁰

In some ways, the contention that the U.S. national security establishment fails to adequately account for the uncertainty of the future seems questionable. After all, as sprawling as it is, the departments and agencies that comprise it house multiple policy offices ostensibly dedicated to considering the long term, such as the State Department’s Policy Planning Staff. They are required by Congress to produce various strategic documents, such as the National Security Strategy, that are intended to account for the long term. And the Pentagon uses scenarios to produce one of the most high-profile of those documents: the National Defense Strategy. Besides, there is the National Intelligence Council’s Global Trends report, an explicit example of strategic foresight that is widely praised within the foresight community.

Yet, in examining these potential objections to this report’s thesis—namely, that the U.S. government suffers short-termism in part because it fails to use strategic foresight—four cautionary themes emerge:

• Short-term demands often crowd out long-term planning, even in units ostensibly dedicated to the latter;
• Strategy documents often fail to affect policy—i.e., they do not link future-thinking to present-doing;
• Contingency planning, which prepares for a single, well-defined future, often substitutes for true scenario planning, which addresses uncertainty; and
• The Global Trends reports are, in many ways, the example that proves the rule, being foresightful but also disconnected from the policymaking process.

These themes are not universal. Any enterprise as complex as the U.S. national security establishment defies easy generalizations. Nevertheless, these dynamics reinforce the case for a whole-of-government approach to strategic foresight.

In the executive branch, the urgent has long been the enemy of the important. Even when it is not battling crisis—and during the 2020 presidential campaign, Joe Biden said that the United States faced no fewer than five crises¹¹—the grind of daily operations in national security tends to crowd out long-range thinking, even though policymakers have considered long-range thinking vital at least since the United States became a global power at the end of World War II.¹²

In 1947, seeing that day-to-day demands were diminishing opportunities to address broader issues, Secretary of State George C. Marshall established the Policy Planning Staff as a discrete office, instructing its inaugural director, George Kennan—who essentially outlined the U.S. Cold War strategy of containment in his “Long Telegram”¹³ and the “X Article”¹⁴—to “avoid trivia,” an injunction that became the office’s motto. In his memoirs, Dean Acheson, who succeeded Marshall as secretary of state, wrote that the purpose of the office was “to look ahead, not into the distant future, but beyond the vision of the operating officers caught in the smoke and crises of current battle; far enough ahead to see the emerging form of things to come and to outline what should be done to meet or anticipate them.”¹⁵

Although this is the very sort of organizational arrangement that supposedly enables an organization to both “exploit” and “explore”¹⁶—that is, to operate and plan simultaneously by separating the one function from the other—Marshall’s experiment did not work. At least not according to Kennan. In 1949, he quit, dubbing Policy Planning “a failure, like all previous attempts to bring order and foresight into the designing of foreign policy by special institutional arrangements.”¹⁷ The reason was that, although separating his office from the “line of command” gave Kennan and his staff the freedom to think, it also deprived them of influence—of the ability to transform their insights into actions.¹⁸

As a result, many of Kennan’s successors have involved themselves (and their staffs) more deeply in the department’s day-to-day operations. How operational Policy Planning is varies, depending on the director and their relationship with the secretary. But, at times, the Policy Planning Staff has found itself signing off on the reams of paper—speeches, talking points, policy directives, etc.—that emanate from the secretary of state’s office. The Policy Planning Staff’s mission remains “to take a longer-term, strategic view of global trends and frame recommendations for the Secretary of State to advance U.S. interests and American values.”¹⁹ That said, Policy Planning staffers and outside observers have continued to note that operations often dominate planning.²⁰ At best, the twin risks that Acheson identified—of being “lured into operations” on the one hand, and of succumbing to “encyclopedism” on the other—continue to stress the staff.²¹

Nor is this problem confined to the State Department. Despite the manifest importance of long-term vision to rational policymaking, operations take precedence within the national security apparatus, both in terms of resources devoted and respect accorded. “Doers” are more likely to be promoted and attain leadership positions than “thinkers.”²² As Aaron Friedberg, a Princeton professor and former adviser to Vice President Dick Cheney, put it:

In this world, the most important people are usually those who are perceived to be most directly involved in the making and implementation of policy on the most pressing issues of the day. Such people are intensely busy with meetings, phone calls, and travel; their focus is on operations rather than planning and on tactics rather than strategy.²³

Richard Haass, himself a “thinker” (he headed Policy Planning under George W. Bush and now runs the Council on Foreign Relations), put it more bluntly: “[A]t the end of the day government is an operational enterprise. It is not a university.”²⁴

That may be, but given that every policy is effectively a prediction, policymaking without a serious attempt to anticipate the range of plausible futures is nonsensical or worse—a situation that has led to exasperation at the tyranny of the day-to-day. In 2016, Julianne Smith, the former deputy national security adviser to then-Vice President Biden and the current U.S. ambassador to NATO, wrote that, because of operational pressures, “the incredibly talented individuals advising the President find it virtually impossible to think strategically.”²⁵ Michèle Flournoy, who served as President Barack Obama’s undersecretary of defense for policy, and Shawn Brimley, who served in the Obama White House, echoed this sentiment: “The reality is that America’s most fundamental deliberations are made in an environment that remains dominated by the needs of the present and the cacophony of current crises.”²⁶

In short, instead of a bureaucracy that links fluid expectations of the future with concrete actions in the present, the national security establishment risks becoming an adhocracy that deals with the future only as it becomes the present.

To counteract the pull of adhocracy, Congress has, at various times, mandated the production of strategic documents. For example, the Goldwater-Nichols Act of 1986, which legislated comprehensive national security reform, required that the White House annually produce a national security strategy, with the idea being that strategic goals would be linked to the budget process, thereby institutionalizing a connection between the future and the present. Similarly, in 1997, Congress required the Department of Defense to submit a Quadrennial Defense Review (QDR)—and, later, the National Defense Strategy (NDS)—that laid out a strategy with an eye toward determining military force requirements and informing the annual budget process.

Emulating the Pentagon’s efforts, other departments with national security responsibilities have conducted quadrennial reviews of their own.²⁷ As a result, between 2010 and 2020 alone, the White House produced three National Security Strategies (2010, 2015, 2017); the Defense Department produced two Quadrennial Defense Reviews (2010, 2014) and a National Defense Strategy (2018); the State Department produced two Quadrennial Diplomacy and Development Reviews (2010, 2015); and the Department of Homeland Security produced two Quadrennial Homeland Security Reviews (2010, 2014).

By some measures, then, Washington’s national security policymakers would seem to be soaked in strategic thinking, suggesting a healthy respect for the long term. Reviewing all these documents and the processes that produced them is beyond the scope of this report. However, the key point is this: at best, their impact on policy is unclear, and at times they may reinforce rather than repair the divide between the future and the present.

A conspicuous example is the National Security Strategy. As Rebecca Friedman Lissner, currently the National Security Council’s director for strategic planning, wrote in 2017, “The NSS is supposed to map out a strategy, but over time, the project has devolved into a rhetorical exercise, characterized by grandiose ambitions and laundry lists of priorities.”²⁸ For grandiose ambitions, few documents top President George W. Bush’s emphasis on “ending tyranny” in the 2006 National Security Strategy.²⁹ And many national security strategies read like a laundry list. For example, President Bill Clinton’s 1999 strategy concluded:

Our international leadership focuses on President Clinton's strategic priorities: efforts to promote peace and security in key regions of the world; to create more jobs and opportunities for Americans through a more open and competitive trading system that also benefits others around the world; to increase cooperation in confronting security threats that threaten our critical infrastructures and our citizens at home and abroad, yet often defy borders and unilateral solutions; to strengthen international arms control and nonproliferation regimes; to protect the environment and the health of our citizens; and to strengthen the intelligence, military, diplomatic and law enforcement tools necessary to meet these challenges.³⁰

A large gap separates such lofty goals from the detail needed to translate them into action. Paul Lettow, who served as the NSC’s senior director for strategic planning from 2007 to 2009, reviewed the national security strategies issued since the beginning of the Cold War, finding that many failed to connect planning with operations.³¹ Contrasting recent efforts with those of President Eisenhower, who emphasized the need to link day-to-day problems with an overarching set of principles, he concluded that recent national security strategies had emphasized style more than substance. As a result, even those who complain about the lack of strategic planning often do not consider the NSS helpful.³² As Lettow wrote, the national security strategy is produced “primarily for public consumption, and mostly disconnected from rigorous planning processes—a cross between a speech and a check-the-box exercise.”³³ Lissner concurred: “Rather than forcing the U.S. government to engage in serious strategic planning, it has become a case study in the failure to do so.”³⁴

The quadrennial departmental reviews also often fail to connect strategy to action. In a comprehensive study of the reports produced by the departments of Defense, State, Homeland Security, and Energy, Jordan Tama, a professor of international relations at American University, concluded that while the four-year exercises were forward-looking, they emphasized vision over action. Some were intended to serve as guides to subsequent reviews that would tackle implementation, but those follow-on efforts rarely happened. As Tama explained, “Strategic planning fatigue often sets in after the completion of a quadrennial review, and the effort to operationalize the review’s ideas is often rushed and far less robust than the review process.”³⁵

Former officials have been particularly withering in their critiques of the QDR. In 2015, Flournoy, who was the principal author of the 1997 QDR, testified to the Senate that strategic planning is essential but that the QDR had become a “glossy coffee table brochure written primarily for outside audiences,” in part because it was publicly released as an unclassified document. She said: “Over the years, the QDR has become a routinized, bottom-up staff exercise that includes hundreds of participants and consumes many thousands of man-hours, rather than a top-down leadership exercise that sets clear priorities, makes hard choices and allocates risk.”³⁶ Defense expert Anthony Cordesman wrote the QDR was “a document decoupled from a real-world force plan, from an honest set of decisions about manpower or procurement.”³⁷ Former Sen. John McCain agreed, writing in December 2017 that “defense strategy documents [had become] increasingly divorced from the strategic realities confronting the United States.”³⁸

That month, Congress charged the Pentagon with producing the National Defense Strategy to address these and other issues, but while the NDS may resolve some of the QDR’s problems—among other things, its contents are largely classified, ostensibly permitting a more honest assessment—it did not address one of the most fundamental: the importance of incorporating uncertainty into national security planning.

The U.S. national security establishment has traditionally failed to adequately account for the uncertainty of the future—a shortcoming that prevents it from generating strategy that will provide advantage over a full range of plausible futures.

This failure may be the most conspicuous and, paradoxically, the most difficult to discern in the Department of Defense. On the one hand, the Pentagon embraces strategic foresight in many initiatives that use scenario planning to explore uncertainty through alternative futures. (See below: Strategic Foresight Within the Pentagon) On the other hand, its principal strategic document, the National Defense Strategy (and, previously, the Quadrennial Defense Review), does not incorporate scenario planning in this way, even though scenarios play a role in its formulation. In these documents, Pentagon leaders have used scenarios less to formulate strategy than to assess the capabilities needed to implement existing strategy in situations they consider most likely. In short, they are doing contingency planning rather than scenario planning.

Pentagon officials regularly acknowledge the uncertainty of the future, suggesting they would benefit greatly from scenario planning, which stretches participants’ imagination by challenging their assumptions and helps them formulate strategy robust to many futures. As defense analyst Michael Fitzsimmons has written, “Scenario planning should be one of the Department of Defense’s (DoD) most important tools for developing strategy under uncertainty.”³⁹ And, in 2002, the department did formalize a process for generating scenarios to inform strategic planning. The problem is that it then used the same process to identify the capabilities the U.S. military would need to prevail in those situations. That calculation required the scenarios to be highly detailed, making them arduous to produce and limiting the number that could reasonably be considered. So, Pentagon leaders would choose a limited set of scenarios based on their understanding of strategic aims and anticipated obstacles to them. A 2019 RAND Corporation study explained the process this way:

Traditionally, the Office of the Secretary of Defense (OSD) identifies its defense strategy and investment priorities. OSD then selects scenarios that reflect the central elements of the anticipated security environment, the chosen defense strategy, and office priorities. The results of the analysis of the selected scenarios inform OSD on the size and mix of forces and capabilities called for to implement the chosen defense strategy within expected fiscal limits.⁴⁰

Put differently, the key difference between the way that the drafters of the QDR/NDS have used scenarios and the way that, say, the Coast Guard conducts scenario planning is that strategy spawns scenarios rather than the other way around.

Admittedly, this seemingly backward process, in which strategy drives scenario selection, is partly a function of logistical and organizational necessity. The Pentagon needs to make minutely detailed decisions about force structure, and that is difficult absent a clear idea of what that force will be used for. Nevertheless, as it exists, the process worsens the very problem strategic foresight is designed to ameliorate: the tendency to make decisions based on prior assumptions about the future absent due consideration of alternatives. Fitzsimmons explained the tension:  

A consensus view of the future is actually vital because you have to make all these choices about policy and programs. But the place where it’s weak is in the robustness of the policy to uncertainty in the future. That’s where everything falls apart. That’s where the weakness is: planning for a singular future versus planning for a range of plausible futures.⁴¹

This might be less concerning if the U.S. military had a better track record of anticipating the next major conflict. But it does not.⁴² To the extent one is eliding uncertainty and instead operating based on assumptions, however well-founded, one is engaging in contingency planning, not scenario planning. That is, one is preparing for a challenge one has already imagined. As one defense expert explained, “Scenarios [for the QDR/NDS] are not a mechanism for preparing for a wide range of possible futures. … It’s more, under the rubric of the possible future, there are various contingencies.”⁴³

Consider the 2018 National Defense Strategy, which showcases a particular vision of the future in which the chief threats to the United States are China and (to a lesser extent) Russia—most notably the danger that China will seize Taiwan or that Russia will invade the Baltics, presenting the United States with a fait accompli to which it cannot respond effectively.⁴⁴ The document itself acknowledges the uncertainty of the future,⁴⁵ but the NDS presents a consensus about the world the U.S. military will face over the long term. To its credit, contrary to many “strategic” documents, the NDS prioritizes threats, acknowledging that the United States will have to make trade-offs in military commitments and spending. But this focus also means that the United States is doubling down in its preparations for a particular future—one in which it must counter the “fait accompli strategy,” most notably vis-à-vis China. As Elbridge Colby, an author of the 2018 NDS, testified to the Senate in 2019, “The NDS is specifically designed to deal with this challenge.”⁴⁶

The point is not that this diagnosis is wrong but rather that, over the long term to which the document refers, the geopolitical situation could change radically. To be sure, it would be foolish to ignore the threat that China seems to pose today, but the country’s ascendance is not assured, nor are its leaders’ goals immutably expansionist.⁴⁷ Rather than address this uncertainty, the report effectively codifies the current conventional wisdom. To the extent that it considers multiple futures, they are variations on a theme: regional aggression by China.⁴⁸ In 2018, Mara Karlin—then a defense expert at the Brookings Institution and currently the assistant secretary of defense for strategies, plans, and capabilities—wrote: “The NDS’s diagnosis of the future security environment is consonant with today’s commonly accepted analysis across the defense community, as is its prescription for operating in it effectively.”⁴⁹ It is a diagnosis that Karlin, who is in charge of the 2022 NDS, has brought to the Pentagon: “I believe that the force planning construct should prioritize and focus on China unless and until the security environment changes dramatically,” she wrote in August 2021.⁵⁰

Pentagon planning for the future is therefore characterized by homogenization rather than imagination. To be sure, the conventional wisdom is often right, extrapolation is often an accurate method of anticipating the short-term future, and many indicators suggest that China is a threat. But, over time, the United States will almost certainly be surprised by a different threat. And it must be prepared to respond with agility.

The Global Trends report, which the National Intelligence Council (NIC) has produced every four years since 1997, is perhaps the closest thing that the United States has to a national foresight document. Other, more predictive products, such as the CIA’s Annual Threat Assessment, analyze short-term dangers. However, when it comes to sketching a range of plausible scenarios about what the long-term future might look like—which is to say, when it comes to producing foresight products aimed at the U.S. government broadly (as opposed to a single department)—there are few, if any, equivalents to Global Trends.

In March 2021, the NIC released Global Trends 2040: A Contested World.⁵¹ The report analyzes potential large-scale changes, running from the highly likely (demographic shifts and climate effects) to the less certain (increasing economic complexity). The report wrestles with the question of how such trends might interact within societies, among states, and in the international system, predicting a “more conflict-prone and volatile geopolitical environment.”⁵² Finally, it presents five alternative far-futures, ranging from “Renaissance of Democracies,” in which the United States leads a wave of economic growth and technological development that rests on strong and open public institutions, to “A World Adrift,” in which the West is locked in competition with China in a largely anarchic international system, leaving global problems like climate change to fester. In short, Global Trends 2040 is an archetypal example of foresight. The School of International Futures, a non-governmental organization that recently published case studies of foresight efforts in eight countries, deemed Global Trends “a bedrock document for American foresight work…used by systems across the world.”⁵³

That said, although officials throughout the U.S. national security establishment are often quick to laud Global Trends, they also tend to downplay its influence—or at least its direct influence—over strategy and policy. The Pentagon, for example, does not incorporate the NIC’s scenarios into its work,⁵⁴ nor does the White House use them to guide policy.⁵⁵ As one defense analyst said, “It does get people thinking. But there’s no evidence to say that it has made a difference to U.S. planning or policy.”⁵⁶ The degree of abstraction—the distance between imagined tomorrows and the demands of today—is too great.

The NIC’s work may be most valuable in encouraging policymakers to consider trends outside their area of expertise and in lifting their gazes to more distant time horizons. For example, late in his second term, President George W. Bush established the National Security Policy Planning Committee to focus on issues that lay “beyond the near term,” to monitor emerging trends, and to examine “plausible, high-impact scenarios.”⁵⁷ Supported by Stephen Hadley, then the national security adviser, the committee consisted of representatives from across the U.S. national security establishment, including the lead author of the 2008 Global Trends report. The committee met twice a month, and its work reportedly encouraged consideration of the long-term future, both by National Security Council staff as well as the high-level officials who read the committee’s products, which included a strategy paper and a set of contingency plans. According to one committee participant, those documents did not directly influence policy, but by emphasizing long-term trends, like demographic shifts and climate change, they helped reorient how policymakers saw future national security challenges.⁵⁸

Because the Global Trends reports, while valuable, do not seem to alter top-level decision-making, the reports have sometimes anticipated developments that policymakers failed to address. After the NIC released Global Trends 2040, which highlights the potential threat from China,⁵⁹ Mathew Burrows, the principal author of three previous reports, wrote, “I wish that the warnings about an independently minded China, particularly in Global Trends, had been heeded a decade or more ago, when there was good reason to worry.”⁶⁰ Although the purpose of scenarios is not to predict the future,⁶¹ this example shows how the Global Trends reports are the exception that proves the rule: The report is the federal government’s most comprehensive strategic foresight exercise—an attempt to deal with the uncertainty of the long-term future by detailing trends and painting a range of plausible futures—but its foresight is a function of its freedom from both operations and planning. It is separated not only from the need to act in the present, but also from the need to develop an “official” view of the future because it does not represent the views of the administration or the Intelligence Community.⁶² It is a thought exercise, not a strategy document.

Global Trends, then, would seem to support the notion that “thinking” comes at the expense of “doing.” To “explore,” the U.S. government not only had to separate thought from action, but it also had to downplay the goal of influencing action. (Global Trends 2040’s modest ambition is to serve as an “analytic framework for policymakers.”⁶³) Yet, as we see in the Coast Guard case, it is possible for future-thought to impact present-day action. Indeed, there are signs that foresight is becoming increasingly connected to policy throughout the U.S. federal government.