An executive order is a directive the President of the United States issues to manage how the federal government operates. It sits next to statutes, regulations, and agency guidance, but it doesn’t need a congressional vote. For people who track the machinery of federal lawmaking and rulemaking—the readers of this blog—the executive order is both a tempting accelerant and a structural risk. When a policy goal stalls in committee or fails a floor vote, the Oval Office can sign a numbered memo and declare the matter settled. Except that settlement is often temporary, legally vulnerable, and procedurally shallow. Understanding why takes a tour through the legal basis, the implementation pipeline, and the quiet institutional costs that pile up every time the pen replaces the gavel.
The Constitutional Skeleton: Where Does the Power Come From?
No clause in the Constitution explicitly says the president can issue executive orders. The power is inferred from Article II, which vests “the executive power” in the president and requires that the laws be faithfully executed. In practice, the legal footing for an executive order usually rests on one of two things: a broad statutory delegation from Congress, or the president’s own constitutional authority over the executive branch. When an order leans on a statute, it’s filling in blanks Congress left. When it leans on inherent authority, it’s navigating a much grayer zone—one that courts have policed with varying enthusiasm since Youngstown Sheet & Tube Co. v. Sawyer in 1952.
That case, which struck down President Truman’s attempt to seize steel mills during a labor dispute, gave us Justice Robert Jackson’s three-part framework. In the first category, the president acts with express or implied congressional authorization; his authority is at its maximum. In the second, he acts in the absence of congressional will; the analysis gets murky. In the third, he acts against the expressed will of Congress; his power is at its “lowest ebb.” Most executive orders that make headlines—on immigration, environmental permitting, or federal spending freezes—operate in that second or third category. That’s precisely where the trouble begins.
The Implementation Gap: An Order Is Not a Regulation
Here’s a procedural truth that gets lost in cable-news coverage: an executive order rarely changes anything on the ground by itself. It’s a directive to the executive branch, not a rule binding the public. For an order to have tangible effect, agencies must translate it into regulations, guidance documents, grant conditions, or enforcement priorities. That translation process is governed by the Administrative Procedure Act (APA), which demands notice-and-comment rulemaking for most substantive rules. An executive order can tell the Environmental Protection Agency to rewrite a power-plant emissions standard, but the EPA must still publish a proposed rule, accept public comments, address them, and survive judicial review. That takes years, not days.
The result is a policy purgatory. The president announces sweeping change on a Tuesday. Agencies scramble to issue interim guidance that may or may not be legally defensible. Regulated parties freeze investment decisions. Litigants file for preliminary injunctions by Friday. For months or years, the “policy” exists only as a press release and a docket number, while the previous administration’s rule—itself perhaps born of an earlier executive order—remains in effect. This isn’t governance; it’s a pendulum swinging in a fog.
The Reversibility Problem: Policy on a Merry-Go-Round
Because an executive order isn’t a statute, the next president can rescind or replace it with equal ease. This creates a policy environment where long-term planning becomes irrational. Consider federal contractors subject to a prevailing-wage executive order. One administration raises the threshold; the next lowers it; the next raises it again. Each change triggers a new rulemaking, new compliance costs, and new litigation. The private sector adapts not by optimizing for the rule, but by hedging against its repeal. That hedging is a deadweight loss—resources spent on lawyers and scenario planning that could have gone into productive investment.
Congress, for all its dysfunction, produces statutes that are sticky. The Affordable Care Act survived dozens of repeal votes and a Supreme Court challenge. The Tax Cuts and Jobs Act of 2017 required a reconciliation process that took months to negotiate and will take another act of Congress to unwind. Executive orders, by contrast, are written on sand. A president who governs primarily by executive order isn’t building a durable legacy; they’re renting policy space by the four-year term, with an option for the next tenant to redecorate immediately.
The Ratchet Effect That Isn’t
Some scholars have argued that executive orders create a one-way ratchet: once an order expands executive authority, it’s hard to shrink back. The historical record is more complicated. Yes, the number of executive orders has generally trended upward since the early 20th century, but the scope of orders oscillates wildly. President Trump’s regulatory freeze orders were largely undone by President Biden’s climate-focused orders, which are now being unwound again. The ratchet, if it exists, isn’t on policy outcomes but on the procedural expectation that presidents should use orders to make policy. That normalization is the real institutional shift—and it’s one that Congress has quietly encouraged by offloading its own legislative responsibilities.
Congressional Complicity: The Delegation Doom Loop
It’s tempting to blame imperial presidents for the rise of executive orders. But the legislative branch has spent decades handing them the keys. Major statutes like the Clean Air Act, the Immigration and Nationality Act, and the National Emergencies Act contain broad delegations of authority that invite executive gap-filling. When Congress writes “the Administrator shall regulate as necessary to protect public health” without defining the standard, it’s effectively writing a blank check. The president then uses an executive order to direct the agency on how to fill in the amount. Congress gets to avoid a tough vote, the president gets to claim credit for action, and the agency gets sued by everyone.
This dynamic is procedurally perverse. The Administrative Procedure Act was designed to ensure that agencies make rules through a transparent, participatory process. But when the rulemaking is driven by an executive order that sets an aggressive deadline or a predetermined outcome, the notice-and-comment process becomes a charade. The agency isn’t genuinely seeking public input; it’s reverse-engineering a rationale to support a decision already made. Courts have started to notice. The Supreme Court’s 2022 decision in West Virginia v. EPA—invoking the “major questions doctrine”—signals that the judiciary is less willing to defer to agencies when the policy stakes are high and the statutory authorization is thin. Executive orders that push agencies into major-question territory are now doubly vulnerable: they can be struck down for exceeding statutory authority, or the agency’s implementing rule can be struck down for lacking clear congressional authorization.
The Budgetary Blind Spot: Spending Power vs. Executive Fiat
One of the most common—and most constitutionally dubious—uses of executive orders is to impound or redirect appropriated funds. The Impoundment Control Act of 1974 was supposed to settle this: the president cannot simply refuse to spend money Congress has appropriated. Yet recent administrations have tested the boundaries by declaring national emergencies to access unspent military construction funds, or by delaying obligated spending while “reviewing” programs. An executive order that halts disbursement of already-appropriated grants may violate the ICA, but the remedy is slow. A lawsuit under the ICA can take months to resolve, and by then the fiscal year may have ended, mooting the dispute. The practical effect is that the president can achieve a temporary impoundment even if the legal theory is weak—another example of process decay.
Budget execution is where the tension between executive orders and the legislative power of the purse becomes most acute. The Office of Management and Budget (OMB) plays a central role here, issuing apportionment decisions that can effectively slow-walk spending. When an executive order directs OMB to “review” certain grant programs and OMB responds by delaying apportionments, the line between lawful review and unlawful impoundment blurs. The Government Accountability Office (GAO) can issue legal opinions, but those opinions aren’t self-executing; they require Congress or the courts to enforce them. In the meantime, the policy is effectively in effect.
Why Not Just Pass a Law? The Procedural Calculus
If executive orders are so fragile, why do presidents keep using them? The answer lies in the procedural asymmetry between Article I and Article II. Passing a law requires bicameral majority votes, overcoming a Senate filibuster (or using reconciliation with its strict limitations), and surviving a presidential veto. An executive order requires a pen. Even when the president’s party controls both chambers, the legislative process is slow, unpredictable, and forces public compromise. An executive order is fast, unilateral, and allows the president to control the narrative. For a White House focused on the next election cycle, the choice is obvious—even if the policy shelf-life is short.
But this calculus has a hidden cost: it atrophies the legislative muscle. When the executive branch becomes the primary policymaker, Congress loses practice in the art of negotiation and compromise. Members can take symbolic votes on bills that will never become law, secure in the knowledge that the real action is happening downtown. The result is a legislature that performs oversight theater while the executive writes the substantive rules. Over time, this erodes the separation of powers not through any single usurpation, but through a slow, mutual abandonment of institutional responsibility.
Judicial Review: The Uneven Backstop
Courts are often invoked as the remedy for executive overreach, but they’re an imperfect backstop. First, litigation is slow. A challenged executive order can remain in effect for years while appeals wind through the D.C. Circuit. Second, the standards of review are deferential in many contexts. The Supreme Court’s Chevron doctrine—recently narrowed but not yet overruled—instructs courts to defer to an agency’s reasonable interpretation of an ambiguous statute. If an executive order directs an agency to adopt a particular interpretation, and the agency does so, a court may uphold it simply because the statute is ambiguous. Third, standing requirements often prevent the most natural plaintiffs—Congress members or state governments—from getting into court at all.
The major-questions doctrine, mentioned earlier, is a partial corrective. It says that for questions of “vast economic and political significance,” agencies must point to clear congressional authorization. But the doctrine is new, its boundaries are fuzzy, and it applies to agency actions, not directly to executive orders. A president can issue an order that pushes the envelope, and the order itself may not be reviewable until an agency acts on it. By then, the political landscape may have shifted. The procedural gap between the stroke of a pen and a final judicial ruling is where executive orders do their most consequential—and often irreversible—work.
FAQ: Executive Orders and the Policy Machinery
Can an executive order create a new criminal offense?
No. Under the Constitution, only Congress can define federal crimes. An executive order can direct the Department of Justice to prioritize enforcement of existing criminal statutes, but it cannot create a new crime or set criminal penalties. Any order that purported to do so would be unconstitutional on its face and would likely be ignored by the courts.
How does an executive order differ from a presidential memorandum?
Both are written instruments by which the president directs executive branch actions. The difference is largely procedural: executive orders must be published in the Federal Register and are numbered, while memoranda are generally not required to be published unless they have “general applicability and legal effect.” In practice, memoranda are often used for more routine, internal management matters, but the line is blurry. Both carry the same legal force when grounded in statutory or constitutional authority.
What happens when an executive order conflicts with a statute?
The statute prevails. Under the Supremacy Clause, the Constitution and federal laws are the “supreme law of the land.” An executive order that contradicts a statute is invalid. However, determining whether a conflict exists can be complex. If the statute is ambiguous, the executive order may be upheld as a reasonable interpretation. If the order directs an agency to act in a way that exceeds the agency’s statutory authority, the agency action—not the order itself—is typically challenged in court. The practical result is that an order can push the boundaries of statutory authority for months or years before a court definitively strikes it down.
How many executive orders are typically issued per president?
The number varies widely. Franklin D. Roosevelt issued 3,721 executive orders over his twelve years in office—the most of any president. In contrast, recent presidents have issued far fewer: Barack Obama issued 276, Donald Trump issued 220, and Joe Biden issued 162 in his first term. The raw count, however, obscures the scope of individual orders. A single order directing a regulatory overhaul can have more impact than a dozen orders renaming federal buildings. The trend is toward fewer but more ambitious orders, which increases the legal and procedural risks described above.
The Long-Term Institutional Cost
What gets lost in the debate over any particular executive order is the cumulative effect on governance. Each order that bypasses the legislative process reinforces the expectation that the president should govern by decree. Each order that gets struck down in court normalizes the idea that the judiciary is a co-policymaker. Each order that triggers a regulatory scramble degrades the quality of agency decision-making. Over decades, these effects compound into a system where policy is made in bursts of unilateral action, interrupted by litigation, and reversed every four to eight years. Stability, predictability, and democratic legitimacy are the casualties.
There’s no quick fix. Congress could reclaim its authority by writing clearer statutes and resisting the temptation to delegate hard choices. Courts could tighten standing rules to allow faster review. Presidents could exercise restraint. But each of these solutions requires an institutional actor to voluntarily give up power—a rare event in Washington. Until then, the executive order will remain what it has become: a policy tool that promises speed but delivers fragility, and that substitutes the appearance of action for the reality of durable governance.
Next in this series: a deep dive into the Congressional Review Act—the legislative tool that can undo agency rules but has its own procedural quirks. Subscribe to follow along.