Every few years, Congress creates a program with a built-in expiration date. The logic seems sound: let's try this, see if it works, and if it doesn't, it'll automatically disappear. Problem solved, right? Except here's the thing—almost none of them actually expire. The "temporary" designation becomes a polite fiction, like calling a house guest "just visiting" after they've been sleeping on your couch for fifteen years.

This pattern repeats so reliably that policy researchers have a name for it: the ratchet effect. Programs click forward but never click back. Understanding why requires looking at what happens after a program launches—the human dynamics, political pressures, and bureaucratic momentum that transform even the most explicitly temporary measure into a permanent fixture of government.

The moment a government program starts distributing benefits, something predictable happens: the people receiving those benefits become very interested in keeping them. This isn't cynical—it's human nature. If you've been getting help with your heating bills for three winters, you've probably adjusted your budget around that assumption. The program isn't abstract policy anymore; it's how you keep your house warm.

What makes this powerful politically is organization. Beneficiaries tend to be concentrated and highly motivated, while taxpayers funding the program are diffuse and barely aware it exists. A farmer receiving crop subsidies will drive to a congressional town hall. A taxpayer paying an extra thirty cents? They don't even notice. This asymmetry creates intense pressure in one direction.

Industry builds up around programs too. The contractors administering benefits, the nonprofits helping people navigate applications, the consultants advising on compliance—they all have livelihoods tied to continuation. When renewal time comes, legislators hear from a coordinated chorus of people whose lives depend on the program. The people who might benefit from eliminating it? They're not in the room. They don't even know the meeting is happening.

Takeaway

Programs don't just serve constituencies—they create them. Every benefit distributed builds a network of people with strong incentives to preserve it.

Sunset clauses were supposed to force hard questions: Did this work? Was the money well spent? Should we keep doing this? In theory, the approaching deadline would trigger rigorous evaluation. In practice, renewal often happens before anyone checks the receipts.

Part of this is timing. Meaningful program evaluation takes years—you need data to accumulate, outcomes to materialize, researchers to analyze results. But legislative calendars don't wait. Renewal decisions frequently come up before the pilot period generates enough evidence to judge. Legislators face a choice between extending based on incomplete information or letting something expire that might be working.

There's also the awkward politics of evaluation. If a program turns out to be ineffective, someone has to explain why they supported it. If it succeeds, opponents look foolish for resisting it. The safest political path is often to skip the evaluation entirely—extend the program with vague language about "continuing progress" and punt the hard questions to a future Congress. Nobody wants to be the person who killed something that might have helped people.

Takeaway

The temporary nature of programs is supposed to create evaluation pressure, but the evaluation often gets skipped because the political costs of finding out outweigh the benefits.

Nothing extends a temporary program quite like a crisis. Economic downturns, natural disasters, public health emergencies—each provides a compelling reason why now isn't the right time to let something expire. The program may not have proven itself, but uncertainty makes elimination feel reckless.

This creates a curious pattern. Programs launched during one crisis get renewed during the next, regardless of whether they addressed either effectively. The emotional logic is powerful: people are struggling, this program exists, therefore we should keep it. The actual connection between the program and the crisis becomes secondary to the appearance of doing something.

The really clever bit is how crises reframe the burden of proof. Normally, a program facing sunset would need to demonstrate results to justify continuation. During an emergency, opponents must prove eliminating it won't cause harm. That's a much harder argument to win when people are scared. And since some kind of crisis is always either happening, just ended, or potentially approaching, there's rarely a "good time" to sunset anything.

Takeaway

Crises shift the political burden from 'prove this works' to 'prove elimination won't hurt'—a much harder standard that tends to favor indefinite continuation.

None of this means sunset clauses are useless—they do force periodic reconsideration, which is more than permanent programs get. But expecting them to automatically cull ineffective programs misunderstands how politics actually works. The forces favoring continuation almost always outweigh the forces favoring expiration.

Better policy design might acknowledge this reality. If you're creating something "temporary," build genuine evaluation requirements with teeth. Fund independent assessment. And maybe accept that once something starts, the default is continuation. The real decision isn't whether to sunset—it's whether to begin at all.