How a Presidential Veto Works, and How Congress Can Override It

A presidential veto is the president’s formal refusal to sign a bill that Congress has passed. It stops the bill from becoming law unless both chambers vote, by a two-thirds margin, to enact it anyway. It is one of the strongest checks the president holds over Congress.

This guide explains how a veto works, the difference between a regular veto and a pocket veto, how Congress overrides one, why veto threats matter and why there is no line-item veto. It draws on the Constitution, Senate and House records and the Congressional Research Service (CRS).

What is a presidential veto?

The veto comes from Article I, Section 7 of the Constitution. According to the National Archives transcription of the Constitution, every bill that passes the House and the Senate must be presented to the president before it becomes law. “If he approve he shall sign it, but if not he shall return it, with his Objections to that House in which it shall have originated.”

According to CRS, all bills and joint resolutions must be presented to the president, with one major exception: proposed constitutional amendments, which need two-thirds approval in each chamber, go directly to the states and never reach the president’s desk.

The House Office of the Historian calls the veto one of the most significant tools a president can use to block legislation, adding that even the threat of a veto can change a bill long before it is presented.

What can the president do when a bill arrives?

The Constitution gives the president 10 days, Sundays excepted, to act on a bill. CRS describes four possible outcomes.

The second row matters. A president cannot kill a bill by ignoring it while Congress is in session. If the 10 days pass without action, the Constitution says it becomes law “in like Manner as if he had signed it.”

How a regular veto works

A regular veto, sometimes called a return veto, happens when the president refuses to sign and sends the bill back to the chamber where it started, along with an explanation. The House historian describes it as a “qualified negative veto,” because Congress gets the final word if it can muster enough votes.

The explanation is known as a veto message, or a memorandum of disapproval. The Constitution requires the receiving chamber to enter the president’s objections in its journal and then “proceed to reconsider” the bill. In 1935, according to the House historian, Franklin D. Roosevelt came to the House chamber to deliver a veto message in person.

George Washington issued the first regular veto on April 5, 1792. Congress did not succeed in overriding a president until March 3, 1845, when it overrode John Tyler’s veto of a Senate bill, S. 66.

What is a pocket veto?

The pocket veto comes from the last line of the same clause. A bill the president has not returned within 10 days becomes law “unless the Congress by their Adjournment prevent its Return, in which Case it shall not be a Law.”

If Congress adjourns and the president simply holds on to the bill, it dies. The House historian calls this an “absolute veto,” because there is nothing to override. According to CRS, the only way around one is to reintroduce the legislation, pass it through both chambers again and send it back. James Madison was the first president to use the pocket veto, in 1812.

Why the pocket veto is contested

The fight is over what counts as an adjournment. The House historian notes that presidents have tried to pocket veto bills during short breaks within and between sessions, and Congress has rejected that use. The legislative branch, backed by modern court rulings, says a pocket veto applies only when Congress has adjourned sine die at the end of a session.

The disputes show up in the official counts. The Senate’s veto table notes that attempted pocket vetoes of two bills in the 101st Congress were disputed, and both bills were enacted into law. House records also list recent cases in which a president withheld a signature during a recess but returned the bill anyway, and the House reconsidered it.

How Congress overrides a veto

To override, both chambers must pass the bill again, “the objections of the President to the contrary notwithstanding,” as the House question is worded. Each chamber needs two-thirds of the members voting, with a quorum present, according to a CRS report on override procedure updated in January 2026. The Constitution requires a vote by yeas and nays, with names entered in the journal, so an override cannot happen by voice vote.

Which chamber votes first?

The chamber where the bill began acts first. The House leads on measures labeled H.R. or H.J. Res., and the Senate leads on S. or S.J. Res. measures. If that first chamber falls short of two-thirds, the bill dies and the other chamber never gets a vote. If it succeeds, the bill and veto message go to the second chamber.

Two-thirds of whom?

The threshold is two-thirds of the members who vote, not two-thirds of every seat. CRS notes that “present” votes are not counted when deciding whether two-thirds voted yes, so attendance can matter.

Does Congress have to vote at all?

No. Nothing in the Constitution requires a direct vote on a vetoed bill. Either chamber can refer it to committee, postpone it or table it, and if no vote happens the measure dies at the end of that Congress. CRS notes that it is not unusual for Congress to make no override attempt when leaders do not believe they have the votes.

House and Senate procedure

In the House, a veto message is a matter of high privilege, usually debated for an hour and voted on shortly after it arrives, and a successful or failed override vote there cannot be reconsidered. The Senate usually acts under a unanimous consent agreement. Because the question is debatable there, ending debate without consent would take cloture, which requires three-fifths of senators.

For the earlier stages of lawmaking, from committee work to final passage, see our step-by-step look at how the Senate passes a bill.

Veto threats and Statements of Administration Policy

According to CRS, a veto threat can push Congress to modify a bill before it reaches the president. Besides public remarks, the White House can issue a Statement of Administration Policy, a written message to Congress, typically released shortly before floor action, laying out the administration’s view of a bill.

The Office of Management and Budget coordinates these statements, and CRS says they are generally the first formal sign that a president intends to veto a bill. A statement may say the president intends to veto, or that agencies or senior advisers would recommend one.

Why there is no line-item veto

The president must take or reject a bill as a whole. Congress tried to change that with the Line Item Veto Act, enacted in April 1996 and effective January 1, 1997. It let the president “cancel in whole” certain amounts of discretionary spending, items of new direct spending and limited tax benefits in laws already signed.

The Supreme Court struck the law down in Clinton v. City of New York, decided June 25, 1998. The Court held that the cancellation procedures violated the Presentment Clause in Article I, Section 7. Justice Stevens’s majority opinion noted that Washington understood the clause as requiring him to “approve all the parts of a Bill, or reject it in toto,” and said Congress cannot alter those procedures without amending the Constitution. For more on the Court’s process, see our guide to how the justices reach their rulings.

Presidential vetoes by the numbers

As of October 2026, the Senate’s summary of bills vetoed and the House historian both count 2,599 vetoes since 1789: 1,533 regular vetoes and 1,066 pocket vetoes. Congress has overridden 112, all of them regular vetoes, since pocket vetoes cannot be overridden. That is about 4 percent of all vetoes and about 7 percent of regular ones.

Roosevelt holds the record for total vetoes, and Andrew Johnson had more vetoes overridden than any other president. Seven presidents, including John Adams and Thomas Jefferson, never vetoed a bill. The Senate table credits no pocket vetoes to any president after Bill Clinton, who had one.

The bottom line

A presidential veto lets the president reject a bill passed by Congress, but it is not always the final word. A regular veto goes back to Congress, which can override it with two-thirds of the members voting in each chamber. A pocket veto, available only when Congress’s adjournment prevents a bill’s return, cannot be overridden at all. And because the Supreme Court rejected the line-item veto, the choice is all or nothing: sign the whole bill, let it become law or send the whole thing back. Because an override needs two-thirds of both chambers, a president needs the backing of just over one-third of the members voting in either one to make a veto stick. Follow the latest clashes between the White House and Congress on the LiveNewsWorld U.S. Politics Live center.

Sources

• National Archives, The Constitution of the United States: A Transcription (Article I, Section 7)
• U.S. Senate, Vetoes, 1789 to Present: Summary of Bills Vetoed
• U.S. House of Representatives, History, Art and Archives, Presidential Vetoes (last updated Jan. 9, 2026)
• Congressional Research Service, Veto Override Procedure in the House and Senate (RS22654, updated Jan. 5, 2026), Regular Vetoes and Pocket Vetoes (RS22188) and The Presidential Veto and Congressional Procedure (RS21750)
• Supreme Court of the United States, Clinton v. City of New York, 524 U.S. 417 (1998), via Legal Information Institute