The Presidential Empire

The Presidential Empire

Whichever way Congress might vote on the NSA, the steady creep of our executive branch is unlikely to be reversed.

The NSA headquarters at Fort Meade, Maryland, circa 1950s. Courtesy of NSA / Wikimedia Commons

Writing in 1973, Arthur Schlesinger Jr. gave a name to a particular kind of presidential power that routinely led the United States into war on the president’s terms instead of Congress’s. He called it, as he did his book, “the imperial presidency.”

Obviously, Schlesinger had Vietnam on his mind—and, in particular, he was thinking about the failures of the Johnson and Nixon administrations to conduct the war with substantial congressional oversight. But as a historian, he also charted the development of an ever-expanding executive branch from the early days of the American republic.

Since then, the phrase has had a strange history. Perhaps the oddest aspect of that history has been the use of the phrase by Republicans to denounce President Obama as an aspiring emperor who routinely appoints “czars” to carry out his policies.

The similarities in language result in part because public figures latch onto a convenient pejorative to describe leaders they disagree with. But while Schlesinger was concerned mostly with the president’s role in war-making, Republicans today invoke the “imperial presidency” mostly (though not entirely) when it comes to the field of domestic policy.

The origin of the imperial presidency in each of these spheres is different, however. The imperial presidency’s place in international affairs will be secure—as long as U.S. foreign policymakers seek to preserve the nation’s hegemonic role in the world—and dangerous, because the stakes, war and peace, are so high. In the domestic arena, the imperial presidency is a response to contemporary political gridlock. But although such gridlock has existed for a while, the imperial exercise of presidential power has not become as deeply embedded in our domestic politics as it has—at least as elite consensus would have it—in our nation’s role in the world beyond. Given that the president’s leading role in domestic affairs has significant legal justification, it also poses less of a threat to the ideal that the actions of the U.S. government are always bound by the Constitution. Exploring the different reasons for the imperial presidency in foreign affairs and domestic policy may help identify future possibilities for more democratic control over policy-making in both domains.

Schlesinger’s imperial president was the leader of what students of the constitutional order call the National Security State—or, in their more conspiratorial moments, the “deep state.” As Stephen Griffin’s recent book Long Wars and the Constitution (2013) shows, the National Security State was created during the Cold War, when a bipartisan consensus agreed that presidential leadership was essential to combat the threat to U.S. interests posed by the Soviet Union, which had its own surveillance apparatus and nuclear weapons.

According to this consensus, only a U.S. national security apparatus under the president’s control could gather intelligence about security threats and develop countermeasures, all of which had to be done with a secrecy that precluded widespread congressional participation. Whistle-blowers were the functional equivalent of spies aiding the U.S.’s adversaries.

Consensus over the National Security State weakened after the collapse of the Soviet Union. But by then it had acquired the kind of institutional form that is difficult to dismantle, even under the best of circumstances. The Central Intelligence Agency, National Security Agency, and the armed forces were potent political actors in Washington by the 1990s. They had many allies in Congress and the news media who still believed that the National Security State was needed to protect American interests.

Then, following September 11, a new sense of existential threat emerged among the American populace. With the rise of international terrorism affecting U.S. interests, the state’s target changed. A new consensus transformed the National Security State into the National Surveillance State. Surveillance was needed to study and act against these dispersed threats, especially because many of them were non-state actors.

Originally, the National Surveillance State focused on actors outside the United States. Terrorism—exemplified by actors like the Irish Republican Army, the Tamil Tigers in Sri Lanka, and a number of Palestinian groups that the State Department listed as terrorists—mostly happened abroad. But the September 11 attacks showed that the “homeland” was vulnerable as well; and so the surveillance state began to focus on a large number of people within the United States.

The legal framework of the surveillance state had two parts. According to the consensus supporting it, the president has the inherent power to guard the United States against “sudden attacks,” a phrase used at the Constitutional Convention in 1787 to explain why the Constitution gave Congress the power to “declare war”—not, as the original version had it, to “make war.” President George W. Bush and his legal advisers had an extremely aggressive and liberal interpretation of this inherent power. They contended not only that the president had the authority to initiate a full-scale war without Congress’s involvement, but also that Congress cannot limit the president’s decisions. They argued that statutes that Congress might enact to regulate the president’s actions—such as limitations on torture used to gain intelligence that might thwart a sudden attack—were unconstitutional.

Obama has been careful to retract Bush’s most aggressive legal positions on presidential unilateralism and congressional power, though he has continued to implement some of Bush’s surveillance policies.

Obama could do this because of the second part of the surveillance state’s legal framework: a group of statutes—enacted by Congress, it is worth emphasizing—that establishes the terms on which the National Surveillance State’s activities, including surveillance but extending well beyond, are to be conducted. These statutes include the Foreign Intelligence Surveillance Act (1978, with later amendments), the U.S.A. Patriot Act (2001), and the Authorization for the Use of Military Force (AUMF) against Al Qaeda, the Taliban, and “associated forces” (2001).

The Obama administration relied, with some reluctance, on the 2001 AUMF to justify its on-going operations in Yemen, Somalia, and, in the past year, against ISIS, on the premise that the operations’ targets are either part of Al Qaeda itself, or that they are armed forces that are Al Qaeda’s co-belligerents. This alone indicates how the statutory framework for the National Surveillance State is jerry-built, assembled out of statutes enacted years ago for other purposes. Notably, Obama’s recently proposed AUMF for ISIS, while limited in some ways (a three-year sunset, a statement that it would not authorize the enduring presence of ground troops), does not include a repeal of the 2001 AUMF, which makes it possible for his successors to return to an interpretation that covers all radical Islamist terrorism.

President Barack Obama and the National Security Staff deliberate over taking action
in Syria, in the Situation Room of the White House, August 30, 2013. Official White
House photo by Pete Souza.

National security professionals generally agree that a comprehensive review, revision, and rationalization of the statutory framework is appropriate, although they also think that the surveillance state can scrape by for quite some time without such a revision. Politics has blocked the revision so far. Republicans in particular have opposed changes that, in their view, cast retrospective doubt on President Bush’s actions. They make their case for opposing such revisions by suggesting it will weaken U.S. security.

Edward Snowden’s revelations brought home the fact that one of the main tasks of the National Surveillance State is indeed surveillance. Much of what Snowden brought to public attention were modernized versions of classical espionage conducted by U.S. spies outside the United States. That sort of espionage was completely consistent with U.S. law. Technology meant, though, that surveillance outside the United States inevitably included information about activities by U.S. citizens both outside the nation’s borders and within them. The statutes creating the framework for this surveillance have provisions aimed at limiting its domestic use to cases with a substantial connection to international terrorism. But, Snowden showed us, those provisions were not fully effective, and the scale of modern surveillance meant that even reasonably effective protections against domestic surveillance still left large numbers of innocent people subject to it.

The consensus around the National Surveillance State is itself an obstacle to the system’s revision and rationalization (note, not really “reform”). The national security community does not really know how much it wants to disclose about the surveillance state’s activities. In addition, revising the statutory framework would require agreement about how extensive congressional oversight should be. The statutes passed by Congress that establish the surveillance state permit an incredible amount of discretion for the president and allow only loose supervision by Congress, and most national security experts believe that this should continue.

Congress may have thought that it limited the imperial presidency that Schlesinger worried about by enacting (over President Richard Nixon’s veto) the War Powers Resolution in 1973. The Resolution requires the president to report to Congress when U.S. troops are introduced into hostilities and to withdraw them after sixty days unless Congress authorizes their continued use. Every president has in fact submitted reports, but only with the caveat that they were acting, as they put it, in a manner consistent with the War Powers Resolution rather than saying that they were submitting required reports.

The sixty-day clock of the War Powers Resolution has generally ensured that presidents act unilaterally only with respect to small-scale actions, and that the larger wars since Vietnam—the two Gulf Wars and the conflict with Al Qaeda—have been undertaken with the approval of both branches. In shorter-term conflicts, presidents have in effect avoided the War Powers Resolution, mostly by creative interpretations of congressional action. President Clinton, for example, took the position that military operations in Kosovo were authorized by an evenly divided vote in the House of Representatives (resulting in no action by Congress as an institution) coupled with congressional authorization for him to spend money on the operations. President Obama offered an extremely thin argument that air operations in Libya were not covered by the War Powers Resolution because there was no significant risk that members of the U.S. armed forces were in physical danger and because the intervention beyond sixty days was meant to be brief.

Where do the courts stand in all this? Basically, nowhere. The Supreme Court has developed a number of doctrines that effectively keep the courts out of this field. The Court used the so-called “standing” doctrine to restrict the ability of people who thought that they might have been the subjects of unlawful surveillance to bring suits. It has a “political questions” doctrine that cautions strongly against judicial intervention in matters involving national security policy, even where the policies are challenged on constitutional grounds. And, even when the Court has formally licensed some judicial supervision of the surveillance state—as it did in connection with detainees at Guantanamo Bay—the lower courts have taken a strongly hands-off position.

Members of Congress from both parties have occasionally grumbled about what presidents have done with the discretion they have been given by the National Surveillance State, but they have done almost nothing to bring that discretion under tighter control. The political consensus over the contours of the National Surveillance State is likely to persist as long as there is a general agreement about the role of the United States in international affairs.

In 2001 then-professor and now Supreme Court justice Elena Kagan published a brilliant article in the Harvard Law Review. The article’s title was “Presidential Administration,” and it identified the most important modern development in the law of the regulatory and administrative state—which is to say, the most important development in recent U.S. law.

Kagan noticed that, beginning with the Reagan administration and accelerating through the end of the century, a lot of policy was being developed directly under the president’s guidance without waiting for Congress to endorse presidential proposals. And then there were the czars. They coordinated action by different agencies dealing with a single subject—drug policy first, then others—and reported directly to the president, who provided more general input.

Presidential Administration, according to Kagan, involved exercises of presidential power, and so in many ways it parallels the role of the imperial presidency in international affairs. Presidents since Reagan have contended that statutes enacted decades ago give them discretion to develop innovative policies that Congress couldn’t enact on its own. They exercise that discretion by directing their subordinates to adopt these policies. This is a departure from the classic Progressive-Era vision of how policy development in the administrative and regulatory state should be made. For the Progressives, experts would develop policy based on their wide knowledge of complex issues, which politicians—the Congress, local governments, various elected officials—would then enact. Presidential Administration, on the other hand, moves politics out of the shadows. It’s also a departure from more recent methods of presidential influence over agencies. For a long time presidents put loyalists in charge of agencies, trusting them to align their agencies with the administration’s priorities. President Reagan (and then his successors) worried that these appointees would “go native”—that is, adopt the perspectives of the agencies themselves, rather than the president’s views. So instead, they appointed a set of czars and also used earlier statutes to enact policies from the White House.

Presidential Administration’s roots were in domestic politics—and it was used as much by conservative presidents as by liberal ones. For generations, presidents had provided policy leadership, developing programs that their supporters then introduced in Congress. But as political scientist Stephen Skowronek showed in The Politics Presidents Make (1997), Reagan discovered that changing policy was made difficult by the entrenched power of interest groups, the bureaucracies they were able to influence, and the congressional committees whose leaders thought that they, not the president, should shape agency decisions. And so Presidential Administration became a way to break this “iron triangle.”

As partisan polarization increased, Presidential Administration became even more attractive. In a divisive political environment, presidents were finding it increasingly difficult to get their legislative agendas enacted. Doing so was not impossible, as George W. Bush showed with No Child Left Behind and even Obama, with the Affordable Care Act. But passing legislation consumed a lot of time and political capital. Presidents, therefore, turned to old statutes. They used their offices to push administrative agencies to develop creative interpretations of statutes enacted decades before. The Supreme Court encouraged administrative “creativity” in what has become its most widely cited opinion, the Chevron U.S.A., Inc. v. Natural Resources Defense Council case (1984). Chevron dealt with administrative agencies and the statutes on which their regulations were based—the Environmental Protection Agency and the Clean Air Act in Chevron and in later cases dealing with regulations addressing global warming. Often those statutes use quite general terms—regulate “to protect the public health with an adequate margin of safety,” for example. The global warming case involved the word “pollutants,” and the EPA eventually decided that carbon emissions counted as a pollutant. Chevron said that when the agency came up with a reasonable interpretation of unclear language, that interpretation had the force of law.

The Clean Air Act was adopted in 1970, before the problem of global warming was understood to be a pressing one, and even though it was amended several times over the next decades, nothing in the amendment really addressed global warming. Chevron meant that the EPA could use an old statute aimed at other problems to deal with global warming, as long as its interpretation was reasonable.

Then there are the czars. Presidents appoint czars to deal with new policy problems that cut across regulatory areas, like managing the recent automobile bailout. In a different political environment, presidents might send legislation to Congress. Believing that to be pointless, however, most presidents have decided to appoint czars to pull together everyone who has existing statutory authority in a particular field of policymaking. The czars have no power to develop new regulations, but their prominence and White House credentials give them enormous influence over those who do the regulatory work—and this helps enact presidential policies without congressional oversight.

Presidential Administration has an additional dimension, illustrated by Obama’s recent decision to suspend deportation of a large number of noncitizens who entered the country without authorization. Modern statutes give presidents and administrative agencies a lot of discretion. That discretion is always exercised for the purpose of advancing some policy, the content of which is sometimes made explicit and is sometimes implicit.

Obama emphasized that Congress provided only enough money to deport 400,000 people a year, which meant that the immigration authorities had to have some priorities. Deporting everyone who came to the authorities’ attention was simply impossible, not to mention unwise and unjust. Obama’s policy makes the reasons supporting discretionary choices clear to the public.

Other statutes, like No Child Left Behind and the Affordable Care Act, contain provisions expressly authorizing the agencies that administer them—which means the president, on issues that matter to him—to waive some or many provisions. Waivers allow the president to bypass congressional decisions that seem to him ill-advised. Sometimes the waivers are small, such as deferring the effective date of some of the Affordable Care Act’s provisions. Sometimes the waivers are what law professor Todd Rakoff and now-judge David Barron call “big waivers.” Big waivers let the administration approve substantial alternatives to the programs Congress enacted, often as an experiment to see whether something out there might work better than what Congress had in mind. The Obama administration has used big waivers in connection with No Child Left Behind, which most experts in the field think was probably badly designed in the first place, and in the Affordable Care Act’s expansion of Medicaid.

As with the National Surveillance State, so with Presidential Administration: presidents ground their actions in existing statutes but they are also bypassing the current Congress and relying on what past Congresses have done. Here too some of their interpretations are quite creative, almost necessarily so in light of the fact that presidents use old statutes to deal with new problems. “Creative”—it should also be noted—doesn’t always mean unlawful.

Courts are more active in examining whether Presidential Administration is lawful than with the National Surveillance State, but here too the legal rules give the president a lot of leeway. The House, for example, has threatened to sue Obama for some of the small ACA waivers that he has enacted during his tenure, but few constitutional scholars think that the suit has any chance of success.

Presidential Administration and the National Surveillance State are almost certainly here to stay because they arise from the persistent quagmire of congressional politics and our hegemonic tendencies in foreign policy. These conditions will change only in the face of new political mobilizations around these two issues. At present it seems to me unlikely that we will see popular mobilizations strong enough to destabilize the consensus around the need for the National Surveillance State. Mobilizing partisans to produce a unified government in which Presidential Administration is replaced by collaboration between the president and Congress seems somewhat more likely. What domestic policies that collaboration would yield depends, of course, on whether Democrats or Republicans are in control.


Mark Tushnet, author of In the Balance: Law and Politics on the Roberts Court (2013), is a professor of law at Harvard Law School.