Clermont, Florida Correspondent — Emergency authority to close for short periods when hospitals are at immediate risk is one thing. Prolonged closure on models carrying wide uncertainty bands is another.

Models are planning tools. When they prove imperfect, and most early ones did, the justification for continuing heavy restrictions weakens rather than staying constant.

My clients are small professional firms and I watched a specific failure that I’d put in the record. The rules distinguished between businesses by category rather than by activity, so a firm of six people in separate offices was treated the same as a crowded room, and a firm that could have operated safely stayed closed while a category that couldn’t stayed open. Nobody could appeal the category.

Short-term powers, yes. Prolonged closure should require tighter evidence, public review, and a sunset.

Dayton, Ohio Correspondent — Temporary closure when a local medical system is near collapse can be justified. I’d defend that without much hesitation.

Extending it for months on models that repeatedly needed major revision is harder. The economic and educational costs compound while forecast precision doesn’t improve.

From the manufacturing side, the thing that struck me was that we were declared essential and we operated the whole time, with distancing and screening and ventilation changes, and our case rates were not the story. That was true across a lot of industrial employers.

If a plant floor could be made to work, the question of why a classroom couldn’t deserved a better answer than it got.

Myrtle Beach, South Carolina Correspondent — Short emergency closures, yes. Months of closures on models that keep changing, no.

Here it was specific and it was brutal. We’re a seasonal economy — a substantial share of the annual revenue arrives in about fourteen weeks. Losing spring wasn’t losing a quarter. For some operators it was losing the year, and the assistance programs were built for businesses with even monthly revenue.

The bar for prolonged measures has to be higher, and part of that bar should be knowing what you’re actually closing.

Sheffield, Jamaica Correspondent — I want to complicate this, because the answers at this table are being given from a country with a great deal of hospital capacity.

Jamaica has a small number of intensive care beds for the whole island. There is no surge to expand into. When the modelling is uncertain in a place like ours, the uncertainty runs in a direction Americans have not had to think about: a modest wave that a wealthy health system absorbs without noticing would exhaust ours entirely, and there is no second hospital to send anyone to.

So a prolonged closure that would be plainly disproportionate in Ohio may not be in Kingston, and we imposed one. It did severe damage — our tourism economy simply stopped — and I would still be reluctant to call it overreach.

Review and sunset provisions are right everywhere. The threshold they enforce cannot be identical everywhere, and I would ask that the American conclusion not be exported as though it were.

Long Island, New York Correspondent — A two-week closure to keep a hospital from collapsing is a government doing a grim job. I’ll defend the spring here, because I watched the refrigerated trucks and I’m not going to pretend that was a modelling artifact.

A season of closures on a model that had already missed, applied to a contractor who can’t work from a screen, is a government that has fallen in love with the model.

Short, local, reviewable — yes. Prolonged on imperfect paper with no automatic expiry — no. An authority that doesn’t expire isn’t public health, it’s a new arrangement of powers that nobody voted for.

What I’d want written before the next one is a statute that says three things. Here is the metric. Here is the date the order dies unless a legislature votes. And here is the distinction between a nursing home and a fifth-grade classroom, which the orders mostly declined to draw.

Las Vegas, Nevada Correspondent — This valley can empty in a weekend and still have the same bills on Monday. A model doesn’t pay the Monday.

A government that closes a property for a quarter on a forecast that will be revised should have to stand inside the revised forecast and say what it got wrong. None of them did.

Authority for a short shock, yes. Authority for a prolonged freeze because the model has become a political comfort object, no.

Write the limit now, while everybody still remembers a banquet room with the chairs stacked on the tables.

Tyler, Texas Correspondent — A diner can’t pivot. A model can. That asymmetry is the whole moral problem and it never appeared in a briefing.

Give a government a short key to slow a surge. Don’t give it a season pass because a slide looked alarming.

Imperfect isn’t an insult aimed at epidemiology. It’s a description of what a forecast is. Descriptions don’t get to padlock a shop in Smith County until autumn.

Legislature, metric, sunset. Without those three, the answer is no.

Jacksonville, Florida Correspondent — Ships run on clocks. Children run on clocks too, and one of those clocks doesn’t have a make-up voyage.

Short authority tied to measured hospital strain. Long authority only if an elected body will put its name on it and stand for reelection under it.

This year inverted that. The people who bore the cost had no vote and the people who imposed it faced no ballot.