The Role of Delegation Theories in Deforming the Constitution

[The third of four blog posts drawing on this week’s Hallows Lecture at Marquette Law School.]

It is not just the “three buckets” picture, in combination with the unitary executive theory, as discussed in my first two blog posts, that has warped the design of the U.S. Constitution. A second “unstated idea,” the topic of this third post, concerns the delegation of authority by Congress.

There is, of course, the longstanding complaint that delegations by Congress should be reined in, so as to force Congress to legislate greater specificity. This has given us agitation, most prominently by Justice Gorsuch, to overturn the longstanding doctrine that delegations are permissible as long as they include an “intelligible principle” guiding the exercise of discretion—the claim being that a more restrictive doctrine is required. More recently, it has given us the major questions doctrine, which says that far-reaching and controversial exercises of delegated authority will be set aside by the courts absent clear authorization by Congress. These are not unstated ideas—they are quite vigorously asserted and debated.

Rather, I am concerned with two other aspects of congressional delegation that are unstated but have contributed greatly to the deformation of our constitutional structure. The first concerns an extremely casual attitude to certain types of delegation to the President or one of the many administrative agencies.

The case against delegation rests on the proposition that the Constitution, in the first sentence of Article I, gives “[a]ll legislative Powers” to Congress. One would therefore assume that sensitivity to delegation would be at its height when the President or some regulatory agency claims the power to issue so-called “legislative rules”—regulations that have a force and effect similar to that of a statute. At one time, the courts were very cautious about such delegations, and said they would refuse to recognize agency rules having the force of law unless they were explicitly authorized by Congress.

More recently, however, the Court has adopted something of the opposite presumption: that any statute that mentions “rules” or “regulations”—even if this could plausibly mean housekeeping or procedural rules—also includes the authority to issue legislative regulations, that is, rules that are functionally equivalent to mini-statutes. This newer presumption, which has never been justified by the Court in any considered decision, has the effect of permitting the transfer of lawmaking authority from Congress (whether this was intended or not) to administrative actors and the President.

As should be obvious, the unstated assumption that any reference to rules means authority to make binding legislative regulations has resulted in an enormous transfer of legal authority from Congress to the Executive.

The other aspect of delegation that remains unstated works in the opposite direction, in a sense: the assumption that Congress cannot delegate to itself or to any entity that Congress controls. The starting point for this development is the famous 1983 case of Immigration and Naturalization Service v. Chadha, which held that the so-called legislative veto is unconstitutional.

The decision includes an elaborate discussion of the Constitution’s provisions for enacting a law: a bill must be adopted by both Houses of Congress in identical form, and must then be presented to the President for signature or veto. The statute at issue in Chadha had followed these procedures; it passed both Houses and was signed by the President. The statute delegated broad authority to the Attorney General to withhold the deportation of undocumented immigrants if certain equitable conditions were met.

But there was a condition: the decision to withhold deportation had to be communicated to the relevant committees of Congress with an explanation, and if either the Senate or the House Committee disagreed with the decision, it could be disapproved. The Court held that the provision allowing a committee to disapprove the decision to withhold deportation was a “legislative act,” because it changed the legal status of persons outside the legislative branch. As such, it violated the provisions of the Constitution spelling out how Congress can enact a statute.

But what about the delegation to the Attorney General? This, too, changed the legal status of persons outside the legislative branch. In a footnote, the Court said that the delegation to the Attorney General to withhold deportation was not a legislative act because . . . well, because this was a decision by the Executive, and each branch is presumed to act in accordance with its designated constitutional authority.

Justice White, in a dissenting opinion, pointed out the question-begging nature of the Court’s analysis. What the decision amounted to, as was confirmed by later decisions, is that Congress may delegate legislative authority to the Executive Branch, but may not delegate legislative authority to itself, a subunit of itself, or a legislative agent. The proposition that a principal, in an otherwise proper act of delegation, can delegate only to a stranger and not to a subordinate is not reflected in general agency law.

Maybe there were sound reasons for the distinction between delegations to executive agencies and delegations to congressional agencies. But if so, they were basically (if I may) unstated.

To make matters worse, Chadha and follow-on decisions declared that any legislative veto could be severed from the statute, and the remainder of the act upheld without it. The effect was to declare unconstitutional some 200 legislative vetoes that had been enacted by Congress as a condition of delegating authority to various executive authorities. Rather than giving Congress an opportunity to reconsider these delegations, the Court decided that the provision for the veto could be severed from the acts, leaving an unconstrained delegation to the Executive.

In one fell swoop, the invalidation of the legislative veto and the severability rulings transferred enormous power from the Congress to the Executive.

Some Supreme Court Justices, at the oral argument and in the recent decision concerning President Trump’s attempt to impose tariffs under the International Economic Emergency Powers Act or IEEPA, belatedly recognized that the Chadha decision has created a one-way ratchet transferring unchecked power to the President. Broad emergency powers like those given in IEEPA were delegated to the President in statutes passed before the Chadha decision, the expectation being that the power would be exercised with discretion, and if not, Congress could override the President’s action with its legislative veto. Now, the only way to rescind these broad delegations is by new legislation, which the President would almost certainly veto.

Chadha‘s unstated principle that Congress cannot delegate to itself has resulted in an enormous transfer of power to the Executive, and a corresponding deformation of the Constitution.

Tomorrow, in my final blog post, I will take up one more unstated idea that has contributed to deforming our constitutional regime. It involves the Supreme Court’s conception of its role.