INS v. Chadha Wiped Out the Legislative Veto in Nearly 200 Laws at Once

On December 16, 1975, the House of Representatives disposed of a man's future in a few seconds. No debate, no recorded vote, no hearing at which he could speak. A resolution passed on a voice vote, and six people whom the Attorney General had decided could stay in the country were told, in effect, to go. One of them was Jagdish Rai Chadha.

Congress had every reason to think it was acting within its rights. It had written the arrangement into the statute itself, and it had done the same in dozens of other laws. Then Chadha sued, and eight years later the Supreme Court held that the whole device was unconstitutional. The ruling erased a tool from nearly 200 statutes at one stroke. It is one of the most consequential separation-of-powers decisions of the twentieth century, and it is also a clean illustration of a principle conservatives defend: the lawmaking procedure in the Constitution is the people's protection, and convenience is not a reason to skip it.

Who was Jagdish Chadha

Chadha was an East Indian man born in Kenya who held a British passport. He came to the United States on a student visa in 1966. His visa expired in 1972, and he stayed.

The Immigration and Naturalization Service began deportation proceedings. Chadha asked for the relief the statute allowed. Section 244(a)(1) of the Immigration and Nationality Act let the Attorney General suspend a deportation if the person had been in the country for seven years, was of good moral character, and would suffer "extreme hardship" if removed. An immigration judge suspended Chadha's deportation in 1974.

Here is where the statute got interesting. Section 244(c)(2) gave Congress a window. After the Attorney General reported a suspension to Congress, either house could adopt a resolution within a set period, disapproving it, and the suspension would be undone. One house. No second chamber, no presidential signature. That is the legislative veto.

The legislative veto, explained

A legislative veto is a clause in a statute that lets Congress, or a part of it, cancel a decision of the executive branch without passing a new law. A typical clause says something like: the President or an agency may act, unless within sixty days either house, or a single committee, passes a resolution disapproving.

Congress wanted to hand power to the executive, and it wanted a leash. A statute that delegates power and attaches a one-house veto seems to give you both the efficiency of delegation and the control of oversight. By the 1980s the legislative veto had spread into nearly 200 statutory provisions, by Justice White's count, on many subjects.

The House Judiciary subcommittee on immigration, chaired at the time by Representative Joshua Eilberg, reviewed the suspension cases. The subcommittee objected to six of them, and Chadha's was among them. The full House, without a debate, passed a resolution vetoing the six, and Chadha was ordered deported.

He went to court. The Ninth Circuit held in 1980 that the one-house veto violated the separation of powers. The Supreme Court took the case, and on June 23, 1983, it ruled for Chadha by a vote of seven to two.

The holding

Chief Justice Warren Burger wrote the majority opinion. His reasoning was less a doctrine than a close reading of two clauses.

The Bicameralism requirement. Article I, Section 1 says: "All legislative Powers herein granted shall be vested in a Congress of the United States, which shall consist of a Senate and House of Representatives." Congress is two houses, and legislative power belongs to the two together. A single house is not "Congress."

The Presentment requirement. Article I, Section 7, Clause 2 describes how a bill becomes law: it passes both houses, and "before it become a Law, [it] shall be presented to the President of the United States." If he signs, it becomes law. If he vetoes, two-thirds of each house may override. Clause 3 closes the loophole: "Every Order, Resolution, or Vote to which the Concurrence of the Senate and House of Representatives may be necessary (except on a question of Adjournment) shall be presented to the President of the United States; and before the Same shall take Effect, shall be approved by him."

Burger's logic ran in steps.

1. Was the House's action legislative in nature? Yes. It had the purpose and effect of altering the legal rights, duties and relations of persons outside the legislative branch, including Chadha, the Attorney General and the immigration authorities. The test is not what Congress calls the act. It is what the act does.
2. If so, it must satisfy the Constitution's requirements for legislation, and it did not. It passed one house, and it was not presented to the President.
3. The framers considered and rejected shortcuts. The Convention had debated whether the executive's veto should be absolute, how the Senate should be formed, and whether Congress should be one body or two. Those were deliberate choices.
4. Convenience is no answer. The opinion acknowledged that the legislative veto was a useful tool for Congress, and said that the choices made at the Convention impose burdens on government that often "seem clumsy, inefficient, even unworkable." But that, Burger wrote in substance, was the price of liberty. The framers had lived under governments that acted arbitrarily and had built a slow, difficult process to prevent it.

The Court treated the veto of one man's case as the exercise of lawmaking power, and held that the Constitution's two safeguards, a second house and the President, had been bypassed. The point is worth a pause. The President's veto and the Senate's concurrence are not courtesies. They are the places where a law can be stopped.

The concurrence and the dissent

Justice Lewis Powell agreed with the result but not the reasoning. He thought the majority swept too broadly. In his view, the House had done something different: it had acted like a court. It had decided that Chadha in particular, and five others, should not receive the relief the law offered. An individual adjudication of that kind is a judicial function, and Congress was not a court. Powell would have decided the case on that narrower ground, and left the question of other forms of legislative veto to be settled when they came up.

Justice Byron White dissented, and his dissent is the best statement of the case for the other side. The legislative veto, White argued, was a practical answer to a real problem. In the modern world Congress cannot write laws specific enough to cover every decision, and it must delegate to agencies. If the only check Congress has on a delegation is to pass a whole new law, which the President can veto, then Congress is at the mercy of the executive. The one-house veto kept the delegations within bounds. And, White said, the majority's opinion swept away much more than it needed to. He noted that its reasoning put in doubt the legislative veto in a very large number of statutes, "nearly 200" of them by his count, covering a wide range of subjects.

Justice William Rehnquist, joined by White, dissented separately on the question of whether the veto clause could be severed from the rest of the statute.

I think White's practical point has weight, and I'll state it fairly: if you give an agency broad power and take away the cheap way to rein it in, you will get a stronger executive branch. That prediction came true. But I also think White's argument proves too much. It is an argument that Congress may rewrite the rules of lawmaking whenever it finds the rules inconvenient. If that were so, the Constitution would not be a constitution.

The strongest objection, answered

Here is the objection at full strength. The Constitution does not mention agencies or administrative law. The modern government was built by statutes that give executives wide discretion. The legislative veto was Congress's attempt to keep a share in the exercise of that discretion. Banning it left Congress with two bad choices: delegate less, which hampers government, or delegate and then live with the result. A court that insists on formalism, the objection goes, ends up giving more power to the executive, which is the opposite of what the framers wanted.

The answer comes in numbered points.

1. The cure for broad delegation is narrower delegation, not a shortcut around lawmaking. The Constitution vests "all legislative powers" in Congress. If Congress finds it cannot supervise what it has handed over, the right response is to hand over less. The doctrine of nondelegation, which asks how much power Congress may transfer at all, is the honest remedy, and The Sick Chicken Case of 1935 and the Doctrine That Could Rein In Agencies explains where it stands. The Court's major-questions doctrine, covered in The Major Questions Doctrine Explained, From Tobacco to the Power Plants, is another form of the same check.
2. A veto without presentment is a power without accountability. The House that vetoed Chadha's relief did it with no recorded vote. The Constitution requires a record: Article I, Section 5 directs that the "Yeas and Nays" of either house be entered on request. The people can punish what they can see. A resolution slipped through on a voice vote is a poor model of democratic control.
3. The President's signature is part of the people's protection. The President is the only officer chosen by the whole nation. The framers made the presentment step a guard against a legislature that would gather every power into its own hands. A single house acting alone, or a committee, is the most concentrated form of that danger.
4. Formalism is how a written constitution survives. If the Court may weigh whether a shortcut is efficient, then the lawmaking clauses are only advice. Burger's opinion took the text at its word, which is what a text-minded reader should want.

There is one more point, which I raise because it complicates the story. The Court's formalism did not fix the problem of the administrative state. It arguably made it worse, because it removed a check on delegation while leaving delegation intact. Conservatives who cheer Chadha ought to be honest that it was a half-victory: it closed the shortcut, but the real cure, a Congress that writes its own laws, is still waiting.

What came after

Congress did not give up its desire to supervise the executive. It found workarounds.

• Informal vetoes. Agencies and committees kept up "committee approval" arrangements, in which an agency agrees to seek a committee's blessing before spending money in certain ways. These have no formal legal force. They rely on the agency's wish to stay on good terms with the people who control its budget. The Court's holding did not make them go away.
• Joint resolutions. Where Congress wanted a veto, it could pass a joint resolution that goes to the President, as the Constitution requires. That works as a real check only when the President is willing to sign it.
• The Congressional Review Act of 1996. This is the most important answer. The Act, passed in 1996 as part of a law on small business regulatory enforcement, requires agencies to submit new rules to Congress. Congress may then pass a joint resolution of disapproval that cancels the rule. And since the resolution is a joint resolution, it goes to the President like any other bill. It satisfies Chadha: bicameralism and presentment are both observed. The cost is that a President who favors the rule will veto the disapproval, so the tool works mainly when the White House and Congress are aligned against a prior administration's rules. It was used successfully only once before 2017, to overturn a workplace ergonomics rule in 2001.

The Congressional Review Act is a lesson in how to answer a constitutional ruling. When the Court told Congress the veto was unconstitutional, Congress did not get around it by trickery. It built a mechanism that passes through the front door, and it did the work of lawmaking. That is what the separation of powers asks of a legislature, and conservatives who care about restraining agencies should know the tool well. For the rules and red tape the agencies produce each year, see the red tape tracker.

The Founding yardstick

Measure the case against the Founders. In Federalist 51, Madison wrote that "ambition must be made to counteract ambition." The structure of lawmaking, two houses and a veto, was built so that no single group could govern alone. In Federalist 73, Hamilton defended the President's qualified negative as a shield for the executive against the legislature and a check on hasty or unjust laws. A legislative veto that allowed one chamber to act alone cut against both. The framers did not make the lawmaking process slow by accident. They made it slow so that bad ideas would die in it.

The Chadha decision is, in this sense, a defense of the principle that the process is the protection. A conservative will recognize the pattern from every other place where the text of the Constitution restrains power: the Appropriations Clause, the Origination Clause, the Take Care Clause. Each is a procedure, and each is a promise.

What it means for a citizen

If you want to hold your representatives to account, Chadha hands you a rule of thumb. When a member of Congress says an agency's action is out of control, ask what law gave the agency the power. Then ask why Congress has not changed the law. The answer will often be that Congress prefers to delegate and complain. The remedy lies in the ordinary legislative process, which means in votes you can see and members you can fire at the next election.

The case also reminds us that rights can hinge on boring clauses. Chadha was a man without a country to go to, and the clause that mattered to him was not a grand declaration. It was the requirement that a law pass two houses and be presented to the President. For more on how the Court reads the lines between the branches, see Federalist 78 Called Courts the Least Dangerous Branch. Was Hamilton Right? and Scalia's Lone Dissent in Morrison v. Olson Saw the Independent Counsel Mess.

The case in brief

1. Section 244(c)(2) let one house of Congress overturn the Attorney General's decision to suspend Jagdish Chadha's deportation, and the House did so in 1975 without debate or a recorded vote.
2. In INS v. Chadha (1983) the Court held seven to two that the one-house veto violated bicameralism (Article I, Section 1) and presentment (Article I, Section 7), even though it was efficient.
3. Powell would have ruled more narrowly on the ground that the House had acted as a court, and White's dissent warned that nearly 200 statutes were in doubt.
4. The practical answers were informal committee arrangements and, above all, the Congressional Review Act of 1996, which sends a disapproval resolution to the President.
5. The fix for broad delegation is narrower delegation and real lawmaking, and the process is the people's protection.

What you can do: look up your own representative's votes on the oversight page and ask the office, in writing, which of the laws it voted for gave an agency power it now complains about.