In 2015, the Army Corps of Engineers and the Environmental Protection Agency jointly issued a rule defining the term “waters of the United States” under the Clean Water Act. The rule, running to 300 pages with thousands of pages of supporting documentation, determined which bodies of water fell under federal jurisdiction and therefore which landowners needed federal permits before altering their property.
Under the rule’s interpretation, the jurisdictional reach extended to certain wetlands, seasonal streams, ditches, and water features that had never previously been understood as federal waters. A farmer in upstate New York who wanted to drain a wet section of a field, a property owner who wanted to fill a low-lying area on their land, a rancher who needed to redirect a seasonal creek, all potentially required a federal permit that could take years to obtain, cost tens of thousands of dollars in legal and consulting fees, and might be denied.
The rule was not passed by Congress. It was issued by two agencies under authority they claimed the Clean Water Act provided, applying their interpretation of statutory language the agencies themselves administered. Congress did not vote on whether seasonal ditches were “waters of the United States.” The agencies decided.
This is the administrative state: binding law, with civil and criminal penalties for violation, made by people no one elected, under authority delegated in language broad enough to justify almost anything, reviewed by courts that for forty years were instructed to defer to the agency’s interpretation of its own power.
What the Necessary and Proper Clause Became
The Anti-Federalists were specific about which provisions of the Constitution alarmed them most. Near the top of every list was the Necessary and Proper Clause, Article I, Section 8’s grant of authority to Congress to make all laws “necessary and proper” for carrying out its enumerated functions.
The clause looks modest. Congress has specific powers, to tax, to regulate commerce, to raise armies, to establish post offices. The Necessary and Proper Clause allows it to make the laws needed to exercise those powers. It sounds like a tool for implementation, not a grant of additional authority.
The Anti-Federalists read it differently. Brutus argued that “necessary and proper” was elastic enough to justify almost any legislation Congress wanted to pass, because any law could be framed as necessary and proper to some enumerated function if the connection was drawn loosely enough. He warned that this elasticity would be the mechanism by which federal power expanded beyond its enumerated limits, not through dramatic seizure but through incremental interpretation, each step justified by the last.
He wrote that the clause would allow Congress to “pass any law which they may think necessary and proper to carry into execution any of the powers of the general government”, and that the word “necessary” would be interpreted not as “indispensable” but as “useful” or “conducive to.” Once that interpretation took hold, there was no meaningful limit.
What Brutus could not have fully anticipated was that Congress would take the next step: not just passing broad laws under the Necessary and Proper Clause, but delegating the power to make binding rules to agencies, creating a class of quasi-legislative bodies operating under congressional authority but without congressional accountability.
The Federal Farmer’s warning extended the concern: a legislature that surrenders its lawmaking function to administrators who are not elected and cannot be removed by the voters has effectively dissolved the representative principle. The people believe they are governed by their representatives. They are actually governed by the administrators their representatives hired and then left unsupervised.
The Scale of What Was Built
The federal administrative state is not a minor appendage to the constitutional structure. It is, by most functional measures, the primary mechanism through which federal power is exercised in daily life.
There are currently approximately 430 federal agencies, departments, commissions, and boards. They employ approximately 2.9 million civilian workers, more than the entire manufacturing workforce of many states. They issue approximately 3,000 to 4,000 new rules per year, each of which has the force of law. In a typical year, for every one page of legislation passed by Congress, federal agencies produce ten to twenty pages of binding regulation.
The Code of Federal Regulations, the compilation of all federal administrative rules, runs to approximately 185,000 pages. No person reads all of it. No person can. The Internal Revenue Code and its accompanying Treasury regulations run to tens of thousands of pages. OSHA’s safety standards, EPA’s environmental regulations, the FDA’s food and drug rules, the USDA’s agricultural standards, the FCC’s communications regulations, each is a body of law larger than the entire legislative output of most state governments, made without a single vote of the legislature.
The people who make these rules are career civil servants and political appointees, selected through processes designed to prioritize expertise and insulation from political pressure. They are not elected. They cannot be removed by voters. They can be removed by the executive branch through a complex process, but in practice career civil servants serve across administrations of both parties, carrying the institutional culture and interpretive preferences of the agencies they inhabit.
The Chevron Era and Its End
From 1984 to 2024, the relationship between the administrative state and the courts was governed by a doctrine called Chevron deference. The doctrine, established in Chevron U.S.A. v. Natural Resources Defense Council, held that when a statute administered by an agency was ambiguous, courts should defer to the agency’s reasonable interpretation of that ambiguity rather than substituting their own judgment.
The practical consequence of Chevron deference: agencies had enormous latitude to expand their authority by interpreting ambiguous statutory language in their favor, and courts were instructed to accept that expansion if it was “reasonable.” An agency that wanted to regulate something not clearly covered by its statute had only to produce a reasonable argument that the statutory language was ambiguous and that its interpretation was plausible. Courts would defer.
Forty years of Chevron deference allowed the administrative state to expand significantly beyond what Congress’s original statutory grants clearly authorized. The “waters of the United States” rule was an example: Congress said the Clean Water Act covered navigable waters, and the agencies interpreted “waters of the United States” to extend far beyond what the statute’s ordinary meaning suggested, and courts deferred.
In 2024, the Supreme Court overturned Chevron in Loper Bright Enterprises v. Raimondo. Courts are no longer instructed to defer to agency interpretations of ambiguous statutes. They must exercise independent judgment.
This is significant, but it does not solve the underlying structural problem. The administrative state still exists. The 185,000 pages of existing regulation still have the force of law. The agencies still issue thousands of new rules each year. What changed is the standard of review when those rules are challenged, not the system that produces them.
The structural problem Brutus identified was not judicial deference. It was the delegation of lawmaking to unelected bodies in the first place. Chevron was a symptom. The delegation is the disease.
What It Costs in Rural Communities
The administrative state’s burden is not evenly distributed. The compliance costs that large corporations treat as a cost of doing business are existential for small operators.
A large agricultural corporation with a legal department and compliance staff can navigate the EPA’s wetlands rules, the USDA’s farm program requirements, the OSHA standards for agricultural workers, and the FDA’s food safety regulations. A family farm with three employees and a credit line stretched across the planting season cannot afford the same legal and consulting infrastructure.
The Federal Farmer warned that governance at a distance from ordinary people would serve organized interests over diffuse ones. The administrative state has produced exactly this distributional consequence: large operators who can afford compliance infrastructure gain relative competitive advantage over small operators who cannot. The regulation that was designed to protect public health or environmental quality ends up functioning as a barrier to entry that benefits the established players and burdens the new entrants and small operators.
The rural small business owner, the independent trucker, the family farmer, the small-town contractor, these are the people most exposed to administrative burden and least equipped to navigate it. They are also the people who had the least voice in the administrative proceedings that produced the rules governing them, because those proceedings happen in Washington and require resources to participate in that most ordinary citizens do not have.
Both Parties’ Administrative State
The administrative state was not built by one party. It was built across administrations of both parties over the course of a century, each adding agencies and authorities to address problems that Congress found easier to delegate than to legislate directly.
Republicans have criticized the regulatory burden on business while expanding national security agencies, adding drug enforcement authorities, and building the homeland security apparatus. Democrats have criticized regulatory capture while expanding the administrative reach of environmental, financial, and labor agencies. Both parties, in power, have found the administrative state useful, it allows policy to be made without requiring congressional majorities that are hard to assemble, and it distributes political responsibility in ways that protect elected officials from accountability for unpopular outcomes.
The structural problem is not that either party favors administrative overreach. It is that both parties find administrative governance convenient when in power, and that the agency bureaucracies persist through administrations of both parties, accumulating authority and institutional culture that neither party fully controls.
The Structural Question
The administrative state exists because Congress, faced with the complexity of governing a modern continental nation, has found it impossible to legislate in sufficient detail and has delegated specificity to agencies. This delegation is not inherently illegitimate, some administrative expertise is necessary and appropriate. The question is where the line is between legitimate delegation and abdication of the legislative function.
The Anti-Federalists’ structural answer was clear: the line is at binding law. Rules with the force of law, carrying penalties for violation, should be made by elected representatives accountable to the people they govern, not by administrators accountable only to the agencies they inhabit. Agencies can advise, implement, and enforce. They should not legislate.
What restoring that line would require: congressional legislation that is specific enough to provide real guidance rather than open-ended delegation, with sunset provisions that force Congress to revisit regulatory authority periodically and confirm or revise it. Agency rules above a certain significance threshold subject to congressional review and affirmative approval before taking effect. A genuine accounting of the regulatory burden on small operators and a structural commitment to calibrating that burden against the operators who bear it.
These reforms have been proposed repeatedly by both parties when in opposition and deprioritized by both when in power. The administrative state is too useful to the party that controls it to be voluntarily restrained.
That structural self-interest is the problem the Federal Farmer identified: governance that controls its own accountability will consistently choose its own convenience over the governed’s consent.
Brutus, Letters I and V, October 18 and December 13, 1787. Federal Farmer, Letter IV, October 12, 1787. Available at the Avalon Project, Yale Law School, and the Library of Congress American Memory collection. Loper Bright Enterprises v. Raimondo, 603 U.S. ___ (2024).