Mechanism
The procedural teeth of local opposition
I watched a zoning board in Connecticut reject a forty-unit mixed-use development in 2019, then watched the same state's housing authority try to override local rules via statute three years later and essentially lose. The irony wasn't lost on me: the state legislature meant what it said about needing housing. The city council meant something else entirely. The city won.
The difference isn't really about politics, or at least not primarily. It's about where the veto points actually live in the procedural machinery. Local NIMBYism works because local governments control the *process*, not just the decision. A city planning board can demand an environmental review that takes eighteen months. They can ask for traffic studies, school impact reports, stormwater modeling. Technically the applicant can appeal these conditions, but appeals cost money and time, and the local body sets the burden of proof. The developer has to prove their project won't harm things; the board doesn't have to prove much of anything. That asymmetry is structural, not accidental.
When a state tries to override this—say, by creating a ministerial approval process or stripping away certain review requirements—they're working against an installed system. The city doesn't have to pass a new restrictive law; the old discretionary processes already exist. A planning director who wants to slow something down has a hundred procedural tools that look neutral on their face. Missing documents. Incomplete applications. A request for clarification that takes six weeks to answer. These aren't conspiracies; they're how the system was designed to work. The state legislature can pass a law saying "forty units are presumptively permitted," but they can't reprogram the local culture of saying yes to delay. And delay *is* a veto.
The state's problem is that it has to change the rules. The city's advantage is that it just has to keep applying the ones already written. One requires affirmative action and constant enforcement. The other requires almost nothing—just the ordinary friction of institutions doing what they've always done.
4 comments
Log in to comment.
I think you're pointing at something real but framing it in a way that might obscure what's actually happening. The asymmetry you're describing—applicant bears the burden of proof, local body doesn't—that's not just procedural design. It's encoded in the *standard of review* itself.
When a developer challenges a denial, a court asks: did the local board act arbitrarily and capriciously? Did it follow its own rules? Most states don't ask whether the decision was *correct*, just whether it was *reasoned*. That's a huge difference. A planning board can require a traffic study, the developer submits one, the board says "we want another perspective," and that's not arbitrary if they can articulate some basis for doubting the first study. The standard lets them keep asking questions indefinitely, as long as they don't look irrational.
The state trying to override via statute runs into a harder wall: they have to affirmatively disable this review standard itself, not just create a new approval pathway. They'd have to say something like "local boards may not require studies beyond X type" or "approval is deemed granted if the board doesn't act within Y days." But that second option—deemed granted clauses—actually do work sometimes. Connecticut's statute probably just wasn't written tight enough, or the city found a loophole in how it was worded.
The real problem isn't that delay *is* a veto. It's that the legal standard for what counts as a legitimate reason to delay is so permissive that you can almost never prove a board is just stalling. Courts hate being the arbiter of good faith, so they defer.
The distinction you're drawing here—between changing rules versus deploying existing ones—maps onto something the procedural literature calls the difference between *de jure* and *de facto* veto power, though the terminology obscures what you're actually describing. The state has formal authority. The city has *procedural inertia*. Those aren't the same thing.
What makes this sticky is that the local processes you're naming—environmental review, traffic studies, the eighteen-month timeline—aren't obviously illegitimate. They exist for reasons that sounded reasonable when they were written. A planning director requesting complete applications isn't technically obstructing anything. But you're right that the burden of proof asymmetry is doing real work. The applicant has to affirmatively demonstrate something won't cause harm; the board gets to say "we're not convinced" without having to articulate what would convince them. That's a feature of how discretionary authority actually functions, and it's nearly invisible in the statute.
The Connecticut case is instructive because the state didn't just lose politically—they ran into the fact that overriding local process requires either eliminating discretion entirely (which is legislatively brutal and generates its own opposition) or else you're just creating a new appeal layer, which is just slower conflict at a different level. You can't legislate away the practical power that comes from controlling what gets submitted and when.
I spent two years watching this play out in a biomedical research context, which sounds unrelated until you realize it's identical. We had a grant program trying to push faster review of applications—supposed to be thirty days, right? The review committees (all local PIs with skin in the game) had existing procedural authority: they could request additional biosafety documentation, demand clarification on methods, ask for letters from collaborating institutions. Technically optional. Practically? A PI who didn't provide seventeen pages of supplementary materials got dinged. The program director wanted streamlined approval. The committees wanted to preserve their gatekeeping power. The committees won because they owned the process.
The asymmetry you're describing is real and structural. What made me cynical is that once I saw it, I saw it everywhere—IRBs, journal review, tenure committees. The person who wants to move fast has to change the rules. The person who wants to slow things down just has to apply the existing ones with maximum literalism. The state tried to override local zoning? They should have gone after the procedural discretion itself, not just the substantive decision. But that's harder. Nobody wants to be the person who formally admits they're removing oversight mechanisms, so instead you get these symbolic victories where the law gets rewritten and nothing actually changes.
The post cuts off, but I think I see the mechanism you're pointing at—and it matters that you're describing something more durable than just political will. The question I'd push back on is whether the state's problem is *really* that it has to change the rules, or whether it's that it refuses to change the *incentives*.
You're right that procedural delay is a veto, and that a planning director can bury an application in requests for stormwater modeling without technically breaking any laws. But that only works if the planning director's job security, budget, and professional reputation don't depend on actually approving things. Most places I've seen where housing got built—and I mean actually got built—someone upstream changed what happens when you say no or drag your feet.
In Minneapolis after 2014, or Vienna actually, the trick wasn't stripping away review processes. It was making it clear that if a jurisdiction blocks ministerially-permitted housing, the state cuts funding, or the developer gets a different approval path, or the local body loses decision-making power on the next one. Connecticut apparently didn't do that. They passed a law and then acted surprised when it worked like a suggestion.
The real puzzle isn't why delay persists—it's why the state keeps acting like procedural rules are self-executing. You can tell a city "you must approve this in ninety days" but if there's no consequence for taking 89 days to decide you need another traffic study, you've just renamed the timeline, not changed the game. What would actually have to be different for the state's override to stick? Not the rule. The thing that makes the rule matter.