Separation of Powers



The framers adopted a system of separated powers to avoid the kind of centralized authority that threatens individual liberty. By separating power among three branches, they hoped that checks and balances would restrain political abuse and preserve republican government. The great bulk of separation of power controversies are handled by the elected branches, not the courts.

Nevertheless, law reviews publish hundreds of articles that examine with microscopic precision judicial rulings on separation of powers. The net result is a mixture of inconsistent and incoherent theories, ranging from pragmatic solutions to purist formulations. How does the federal government function in the face of this doctrinal confusion? The short answer is that government does fairly well because most of the principal disputes involving separation of powers are settled outside the courts. These collisions rarely reach the courts and, if they do, are likely to be pushed back to the elected branches for final resolution. Complex and delicate arrangements are fashioned regularly outside the courtroom, forcing legislators and executive officials to discover mid-ground solutions that meet the needs of both institutions.

Under the best of conditions, the Supreme Court offers limited help in resolving basic disputes over separation of powers. There are simply too many conflicts about issues that are not easily addressed in court. During recent decades, the Court has slipped back and forth in its search for principles, sometimes embracing a functional and pragmatic approach of overlapping powers, only to switch later to a strict, formalistic model of powers cleanly separated. With this confusion, the executive and legislative branches operate under unusual pressure to fend for themselves.

Jurisdictional Barriers: Covert Spending

Federal courts use a number of doctrines to sidestep separation of powers disputes. Through such threshold tests as standing, courts play virtually no role in defining the allocation of power between Congress and the President on issues like the Statement and Account Clause, impeachment, and the Incompatibility and Ineligibility Clauses.

The Statement and Account Clause provides that a “regular Statement and Account of the Receipts and Expenditures of all public Money shall be published from time to time.”1 The phrase “from time to time” suggests that the framers allowed for some secrecy—not total secrecy to keep the public forever in the dark, but limited secrecy. Beginning with the Central Intelligence Act of 1949, Congress chose to rely on secret spending without ever giving an accounting to the public. Congress used covert methods to provide the Central Intelligence Agency, the National Security Agency, and other parts of the intelligence establishment with tens of billions of dollars each year. The money was hidden in appropriations accounts, with no public accounting.

When the issue of covert spending was litigated, federal judges made it clear that the meaning of the Statement and Account Clause would be left to elected officials, not to the courts. In 1974, a federal taxpayer’s effort to have the CIA budget disclosed failed before the Supreme Court. Without disputing the lower court’s assertion that the framers “deemed fiscal information essential if the electorate was to exercise any control over its representatives and meet their new responsibilities as citizens of the Republic,”2 the Supreme Court ruled that the taxpayer lacked standing to sue.3 In a dissent, Justice Douglas expressed surprise that the Court would refuse to adjudicate the case and toss it back to the political branches: “Congress of course has discretion; but to say that it has the power to read the clause out of the Constitution when it comes to one or two or three agencies is astounding.”4

Another private citizen attempted through the Freedom of Information Act (FOIA) to gain access to CIA documents detailing legal bills and fee arrangements of private attorneys retained by the Agency. The effort was unsuccessful because a federal appellate court decided in 1980 that the CIA documents were exempt from disclosure under FOIA. When the plaintiff tried to argue that the FOIA exemption violated the Statement and Account Clause, the court ruled that the plaintiff lacked standing to raise this constitutional challenge against secret appropriations and expenditures for the CIA.5

Thus, the meaning of the Statement and Account Clause depended on Congress and executive officials to flesh out this part of the Constitution. In the 1974 case of United States v. Richardson, the Court said that the subject matter “is committed to the surveillance of Congress, and ultimately to the political process.”6 In a case brought by a member of Congress, a federal appellate court noted that the Statement and Account Clause “is not self-defining and Congress has plenary power to give meaning to the provision.”7

In 1997, under pressure of a lawsuit, the CIA released a figure of $26.6 billion for the budget of the intelligence community. Of that amount, the CIA portion was about $3 billion. In 1998, the CIA voluntarily released a figure of $26.7 billion for the community’s budget, but in 1999 it again refused to disclose the aggregate budget. A lawsuit to obtain the budget total by invoking the Freedom of Information Act was dismissed in 1999.8

In 2004, the 9/11 Commission recommended that the aggregate budget of the intelligence community be made public. Three years later, Congress passed legislation to implement that proposal. The statute required the Director of National Intelligence (DNI) to disclose the aggregate budget, but by fiscal 2009, the President would be authorized to waive the disclosure requirement.9 There has been no presidential waiver. On October 30, 2007, DNI Mike McConnell released the aggregate amount of $43.5 billion for fiscal 2007.10 His figure understated the total. News reports explained that his aggregate excluded funds for the intelligence operations of the military services. Including those amounts, the total substantially exceeded $50 billion.11 The current total ranges between $70 billion and $80 billion.12


The Constitution provides that the President, Vice President, and all civil officers of the United States shall be removed from office upon “Impeachment for, and Conviction of, Treason, Bribery, or other high Crimes and Misdemeanors.” The House impeaches by a majority vote; a two-thirds vote of the Senate is needed for conviction. Other than making the Chief Justice the presiding officer in the Senate after a President has been impeached, the framers excluded the judiciary, and the courts have not tried to carve out a role. In 1993, the Supreme Court turned aside a challenge to the impeachment of federal judge Walter Nixon, dismissing the case as a nonjusticiable political question.13

Legislators decide many key issues, such as the burden of proof. Should it be clear and convincing evidence? A preponderance of the evidence? Beyond a reasonable doubt? Lawmakers are at liberty to select whatever test they are comfortable with. The two chambers, moreover, determine crucial procedural issues, including the right of the accused to confront and cross-examine witnesses and the opportunity of House managers to call witnesses. Legislators make judgments to accept or exclude evidence and determine issues of relevance and materiality. Senators decide whether deliberations at the end of the trial should be open or closed.

What are the grounds for impeachment? Treason is defined in Article III, Section 3 of the Constitution, while bribery is generally understood to mean the giving, offering, or taking of rewards as payment for favors. However, there is continuing disagreement about the meaning of “other high Crimes and Misdemeanors.” Does the word “crimes” apply only to actions indictable in the courts (statutory offenses), or did the framers have in mind something broader, covering abuse of office and “political crimes” against government and society? “Other” suggests that high crimes and misdemeanors are of the same order as treason and bribery, but precisely what this clause means is left to the individual interpretations of lawmakers, not courts.

During the impeachment of President Richard Nixon, House Judiciary Committee staff concluded that impeachment “is a constitutional remedy addressed to serious offenses against the system of government,” and, consequently, criminality was not essential to impeach and remove a federal official.14 Members of Congress may conclude that a federal official, although guilty of a crime, ought not to be removed from office. During the vote on the removal of President Bill Clinton, for example, several Senators announced that he was guilty as charged (of perjury and obstruction of justice) but that the nature of the offenses did not justify removal.15

Clinton’s impeachment is instructive for other reasons. Throughout the process, several members hinged their votes on matters having nothing to do with the specific allegations made against the President. Democrats, for example, spoke about the wrongness of “overturn[ing] the results of a national election” and the need to “[l]‌isten to the people of America. They do not believe impeachment is a proper remedy for President Clinton’s misbehavior.”16

Some lawmakers suggested a congressional resolution of censure as a substitute for impeachment and removal. Although Congress is authorized by the Constitution to censure its members for “disorderly behavior,” the act of censuring officials outside the legislative branch raises serious questions of separation of powers.

Some guidance on this issue comes from the Senate’s adoption of a resolution in 1834 censuring President Andrew Jackson for assuming “authority and power not conferred by the Constitution, but in derogation of both.” This resolution was a response to Jackson’s decision to remove a Secretary of the Treasury for refusing to transfer the government’s deposits from the central U.S. Bank to state banks. Jackson, outraged by the censure resolution, complained that “without notice, unheard and untried, I thus find myself charged on the records of the Senate, and in a form hitherto unknown in our history, with the high crime of violating the laws and Constitution of my country.”17 Jackson argued that if Congress wanted to charge him with constitutional violations, it had to act through the impeachment process. Three years later, the Senate ordered the censure resolution expunged from its record.18

In December 1998, during the impeachment debate on Clinton, the House prevented a vote on a censure resolution offered by Democrats. Echoing Jackson’s claims, Speaker-elect Bob Livingston (R-La.) said that censure “is out of the realm of responsibility of the House of Representatives. We have a constitutional responsibility to charge or not charge, impeach or not impeach.”19 Senate efforts to adopt a censure resolution also failed. With the rejection of those resolutions, censure no longer appears a viable substitute for impeachment.

Incompatibility and Ineligibility Clauses

The framers did not intend a pure separation of powers, with each branch of the national government hermetically sealed and cut off from other branches. However, they added to the Constitution two provisions designed to keep the executive and legislative branches at a certain distance. The Constitution prohibits members of either house from holding any other civil office (the Incompatibility Clause) and prohibits members of Congress from being appointed to any federal position whose salary has been increased during their term of office (the Ineligibility Clause).20 Judicial reliance on jurisdictional barriers has ensured that the meaning of those two clauses has been developed almost entirely by the elected branches.

The framers included the Incompatibility and Ineligibility clauses to prevent the executive from using the appointment power to corrupt legislators.21 They knew the English Crown had used appointments to undermine the independence of Parliament.22 The Incompatibility Clause has existed for two centuries without any definition or application by federal courts. When the clause reached a district court in 1971, in a case involving the right of members of Congress to hold a commission in the armed forces reserves, the judge remarked that the “meaning and effect of this constitutional provision have never before been determined by a court.”23

Three years later, the Supreme Court held that the plaintiffs lacked standing to bring their case.24 In response to the objection that if courts fail to resolve the issue of the incompatibility clause, then as a practical matter no one can, the Court replied: “Our system of government leaves many crucial decisions to the political processes.”25 In 1977, when the Justice Department examined the issue of whether members of Congress may hold commissions as officers in the armed forces reserves, it concluded that the “exclusive responsibility for interpreting and enforcing the Incompatibility Clause rests with Congress.”26

As to the Ineligibility Clause, interpretations by Congress and the executive branch have far outweighed contributions from the courts. Opinions by attorneys general from 1882 to 1895 held that members of Congress were ineligible under the Constitution to accept appointment to an executive branch position.27 Later, however, the executive branch reached a settlement with Congress to nominate a member of Congress who was ineligible under a literal reading of the Constitution. In 1909, President William Howard Taft wanted Senator Philander Knox to serve as Secretary of State, even though the salary of that office had been increased during Knox’s term. As a way of removing part of the constitutional problem, the Senate passed legislation to reduce the compensation of the Secretary of State to the previous level.28 That did not satisfy the express language of the Ineligibility Clause, but it appeared to take away the appearance of gain and corruption.

After the bill passed the Senate without debate and without a recorded vote, substantial opposition developed in the House. Congressman James B. (Champ) Clark, who would serve as Speaker from 1911 to 1919, strongly objected to the nomination: “[W]‌e all know that this bill is an attempt to make a man eligible as Secretary of State who is ineligible under the Constitution of the United States. [Applause.] This bill is simply an effort to override the Constitution by statute …. It is a question of the construction of the Constitution. It is a question of understanding plain English….”29

Congressmen Oscar W. Gillespie and Edwin Y. Yates agreed, insisting that the provisions of the Constitution in question “are plain, they are emphatic, they are unequivocal” and “it is clear to even a layman as to what the clause in the Constitution says and means.” Nevertheless, the House passed the bill by the vote of 173 to 116, largely on the ground that the Presidents has a right to select who he wants for the Cabinet and that the bill satisfied the spirit of the Ineligibility Clause.30 The bill, providing for the repeal of the increase in salary for the Secretary of State, was enacted. 31

The Ineligibility Clause has presented a challenge to the appointment of some judges. Senator Hugo Black was nominated to the Supreme Court in 1937, although a retirement system for the judiciary had been enacted that year while Black served in the Senate. In response to a case raising the constitutional issue, the Court held that the plaintiff lacked standing to bring the suit.32 In 1979, the nomination and confirmation of Congressman Abner Mikva to the D.C. Circuit was challenged on the ground that the salaries of federal judges had been increased during his term in Congress. Once again, the suit was tossed out because of lack of standing.33

In 1973, President Nixon wanted to nominate Senator William Saxbe to be Attorney General, even though the salary of that office had been increased during Saxbe’s term as Senator. The Justice Department concluded that Saxbe would be eligible if Congress passed legislation setting his salary for Attorney General at the level established before the increase.34 After lengthy and contentious debate, the bill passed the Senate, 75 to 16.35 With less debate, the House passed the bill 261 to 129, and it became law.36

Similarly, Congress passed legislation in 1980 to permit Senator Ed Muskie to become Secretary of State in the Carter administration. The salary of that office was reduced to its previous level.37 In 1993, to allow President Clinton to select Senator Lloyd Bentsen as Secretary of the Treasury, Congress passed comparable legislation.38 Although subject to academic attack—including an article titled “Is Lloyd Bentsen Unconstitutional?”39—standing barriers prevented a successful lawsuit challenging his appointment. Congress passed legislation in 2008 to permit Senator Hillary Clinton to serve as Secretary of State in the Obama administration.40 A foreign service officer went to court to challenge her appointment as a violation of the Ineligibility Clause. The case was dismissed for lack of standing.41

The Veto Power

If the President decides to withhold his signature from a bill, he is directed by Article I, Section 7 of the Constitution to return it “with his Objections to that House in which it shall have originated, who shall enter the Objections at large on their Journal, and proceed to reconsider it.” If “two thirds of that house” agree to pass the bill, it is reconsidered by the other house and if approved by two-thirds of that house, “it shall become a Law.” Some ambiguities of this language were first clarified by congressional precedents. Those legislative judgments were later sustained by the courts.

Must Congress immediately proceed to reconsider a veto? Initially, that practice was followed under the early Presidents when vetoes were rare. President George Washington exercised his first veto on April 5, 1792. The House of Representatives resolved that the bill be reconsidered the next day, when it sustained the veto.42 Washington’s second veto, on February 28, 1797, was sustained a day later, on March 1.43

There were no vetoes by Presidents John Adams or Thomas Jefferson. When President Madison vetoed a bill on February 21, 1811, members of the House of Representatives debated at great length the propriety of referring a veto message to a select committee for initial review. Some believed the Constitution required immediate consideration. Others insisted that each house had a right to refer a vetoed bill to a select committee for closer study. The override effort took place two days later, with the House sustaining the veto.44

President Andrew Jackson’s fourth regular veto set the precedent for no action at all by Congress: the veto was unchallenged. The Senate concluded that the constitutional requirement of “proceed to reconsider” was satisfied by laying the veto message on the table without moving to either a debate or a vote.45 Jackson sent his last regular veto to the Senate on Friday, June 10, 1836. An unsuccessful effort to override did not occur until 14 days later (Sundays excluded).46

Later vetoes experienced even longer delays.47 Today it is established practice that if the President vetoes a bill, Congress may schedule an override vote at any time during the two years of a Congress. This constitutional question is not litigated. It is left to the rules and procedures of the two houses of Congress.

What is meant by “two thirds of that House” for an override vote? Is it two-thirds of the total membership of each house, or merely two-thirds of a majority present? The House of Representatives early decided on two-thirds of the members present, provided they formed a quorum.48 That ruling was liberalized in 1912 when Speaker Champ Clark announced that an override required two-thirds of the members present and voting.49 On this particular override attempt, there were 174 yeas and 80 nays, with 10 members voting present. Although the 174 fell short of two-thirds of the 264 present, it did constitute two-thirds of the 254 voting. The override therefore carried.50 Building on these legislative precedents, in 1919, the Supreme Court decided that two-thirds of a quorum sufficed for an override.51

May a President sign a bill after the final adjournment of a Congress? For much of U.S. history, Presidents believed they were a constituent part of Congress with respect to lawmaking and therefore could sign legislation only while Congress remained in session. Consistent with that belief, Presidents would come to a special room in the Capitol and sign hundreds of bills in the final days of a Congress. President Grover Cleveland challenged that practice by refusing to go to the Capitol, but relented a year later upon the advice of his Attorney General.52 In 1920 and again in 1931, two attorneys general argued that the President had constitutional authority to sign a bill after the final adjournment of Congress.53 In 1932, the Supreme Court agreed with that assessment.54 Once again, the political branches paved the way for interpreting constitutional language.

Presidents typically exercise two types of vetoes: the regular (or return) veto and the pocket veto. For regular vetoes, after a bill passes Congress, the President may return it with his objections, requiring two-thirds of each chamber for an override. That type of veto power is therefore qualified or conditional. In the case of pocket vetoes, however, the veto power is absolute whenever a congressional adjournment “prevents” the return of a bill.55 Although some court rulings have established important parameters for the pocket veto, there has not been a definitive judicial ruling to clarify the President’s power. As a result, the scope of the pocket veto has been left largely to practice and political understandings developed by the executive and legislative branches.

The Supreme Court did not decide a pocket veto dispute until 1929, when it held that a five-month adjournment of Congress in 1926 “prevented” the President from returning a bill to Congress.56 In 1938, the Court held that a brief recess of three days by the Senate did not prevent the President from returning a bill to Congress, particularly when Congress had authorized legislative agents to receive veto messages during the recess.57 There were no further decisions by federal courts on veto issues until the early 1970s, when Senator Ted Kennedy successfully challenged a pocket veto by President Nixon during a short adjournment.58 The Senate was absent for four days and the House for five. Unlike the 1929 case, Nixon’s action involved a short adjournment during a session rather than a lengthy adjournment at the end of a session.

As a result of this litigation, Presidents Ford and Carter entered into an accommodation with Congress, pledging not to use the pocket veto except at the end of a Congress. They would not use the pocket veto in the middle of a session or between the first and second sessions. Several actions by President Reagan reopened the pocket veto controversy when he experimented with pocket vetoes between the first and second sessions. One of these pocket vetoes was challenged by Congressmen Mike Barnes, who lost in district court but prevailed in the D.C. Circuit.59 Just when it appeared that the pocket veto issue might be resolved judicially, in 1987, the Supreme Court held that the dispute was moot because the bill in question had expired by its own terms.60

The scope of the pocket veto now returned to the elected branches. From 1989 to 2001, Presidents George Bush and Bill Clinton exercised a pocket veto between sessions or in the middle of a session, but instead of treating the “pocket veto” as an absolute veto, they returned the bill to Congress for a possible override. They thus created a third variant, or hybrid: a returned pocket veto! For example, on August 5, 2000, President Clinton vetoed the Marriage Tax Relief Reconciliation Act, claiming that the summer break of Congress (from July 27 to September 5) “prevented my return” of the bill within the meaning of the Constitution as interpreted by The Pocket Veto Case of 1929. Yet he added this qualification: “In addition to withholding my signature and thereby invoking my constitutional power to ‘pocket veto’ bills during an adjournment of the Congress, to avoid litigation, I am also sending H.R. 4810 to the House of Representatives with my objections, to leave no possible doubt that I have vetoed the measure.”61 Similarly, on August 31, he “pocket vetoed” the Death Tax Elimination Act and appended the same explanation.62

On September 6, the House treated the two veto messages as return vetoes, not pocket vetoes.63 On the following day, the House voted 274 to 157 to override the veto of the death tax bill, short of the two-thirds needed.64 During the debate, no one referred to Clinton’s action as a pocket veto. On September 13, the House took up the veto of the marriage tax bill, and again no member regarded Clinton’s disapproval memo as a pocket veto. The vote of 270 to 158 fell short of the necessary two-thirds.65

In his two terms as President, George W. Bush vetoed only 12 bills, with one being a pocket veto that he returned to Congress on December 28, 2007. Congress did not challenge his action.66 However, a month later, it enacted the bill with one change, allowing the President to waive certain provisions regarding the rights of American citizens to sue Iraq for recovery of damages related to torture at the hands of the Iraqi army in 1991.67 In litigation that reached the Supreme Court in 2009, there was a possibility for the Court to judge whether Bush’s return pocket veto was defective. The Court chose not to decide that issue: “We need not inquire into that point, since Congress (evidently thinking the veto effective) enacted a new bill that was identical in all material respects but for the addition of Presidential waiver authority.”68

President Barack Obama engaged in some return pocket vetoes. On December 30, 2009, he signed a “memorandum of disapproval” (signaling a pocket veto) but added: “To leave no doubt that the bill is being vetoed as unnecessary legislation, in addition to withholding my signature, I am also returning H.J. Res. 64 to the Clerk of the House of Representatives, along with this Memorandum of Disapproval.”69 Congress treated it as a regular veto, subject to override.70 Another hybrid veto appeared on October 8, 2010.71 The House, after receiving the bill (H.R. 3808), took a vote to override the veto, but it failed, 185 to 235.72 This odd arrangement of returned pocket vetoes reflects executive-legislative accommodations.

Recess Appointments

Congressional recesses and adjournments invite another separation of power disagreement: the authority of the President to make recess appointments. The framers realized the Senate would not always be in session to give its advice and consent to presidential nominations. To cover those periods of absence, the President “shall have Power to fill up all Vacancies that may happen during the Recess of the Senate, by granting Commissions which shall expire at the End of their next Session.”73 A determination by the President to exploit this power to the fullest would undermine the Senate’s constitutional power over confirmations.

The reach of the power to make recess appointments has been defined primarily by the legislative and executive branches, not by the courts. An early issue involved the meaning of “happen.” Does that mean only vacancies that “happen to take place” during a recess, or is it the broader reading of any vacancy that may “happen to exist” at the time of a recess? A long list of opinions by Attorneys General favored the latter interpretation.74 Those opinions opened the door to possible abuse by the President, relying on recess appointments instead of the regular process requiring Senate approval. In the face of these executive interpretations, Congress responded with statutory restrictions.

In a report issued in 1863, the Senate Judiciary Committee rejected the opinions of Attorneys General. Interpreting the constitutional language, “may happen during the Recess of the Senate,” to include what happened before the recess seemed to the committee “a perversion of language.”75

Only gold members can continue reading. Log In or Register to continue