“Pay your fair share.” You’ve heard it at the Statehouse, on the campaign trail, in the comments under every property tax story we’ve run. It sounds like arithmetic. It isn’t. It’s three different claims wearing one slogan, and the reason the argument never ends is that none of the three ever names a number.

So we built something that does. The Fair-Share-O-Meter takes a Vermont household, runs it through the state’s actual 2025 tax code, the FY2027 budget, and the Tax Department’s own return data, and shows where that household stands under each definition of “fair.” It doesn’t pick one. That’s the point.

Three definitions, one word

Equal share. Everyone pays the same dollars. Vermont’s budget this year is $9.38 billion, about 35 percent of it federal money. The remaining $6.1 billion is what Vermont taxes actually cover. Divide it by 644,663 Vermonters and an equal share is $9,458 a person. A family of four owes $37,800 under this definition. Almost nobody pays it. Almost nobody says it out loud either, but it is one honest meaning of the words.

Proportional. Everyone pays the same percentage of income. Spread the same $6.1 billion across the $30.8 billion of adjusted gross income Vermonters reported in 2024 and the flat rate is 19.8 percent. On the broader federal measure of personal income it’s about 12 percent. Either way, the bottom third of returns, which currently gets a net refund from the state income tax, starts getting a bill.

As enacted. Ability to pay, as the Legislature wrote it. This is the lens the slogan usually means, and it’s the only one that never says how much. It says “more.”

What “more” already looks like

The Tax Department’s 2024 percentile table is the cleanest picture of who pays Vermont’s income tax, and it’s worth reading before anyone argues about it.

The top 1 percent of resident returns, 3,298 of them, start at $586,634 of AGI. They earned 16.7 percent of all Vermont AGI and paid 29.8 percent of all net Vermont income tax. Their average return paid $104,883. The top 10 percent earned 43.3 percent of the income and paid 64.3 percent of the tax. The bottom half earned 13.8 percent and paid 2.6 percent. Below the 30th percentile the number goes negative: refundable credits exceed what’s owed.

Tax share divided by income share is a quick way to read a progressive system. For the top 1 percent in Vermont it’s 1.78. For the bottom half it’s 0.19.

That is the income tax alone. It is about a quarter of what the state raises and the only steeply progressive piece. Add sales, excise and property taxes and the Institute on Taxation and Economic Policy’s figures flatten the picture considerably: from roughly the 40th percentile up, every income group in Vermont pays between 9.6 and 10.6 percent of its income in state and local taxes. The top 1 percent pays 10.1. The middle fifth pays 9.6. On the whole system, Vermont is close to a flat-rate state with a progressive income tax bolted to the front.

None of that tells you whether the top pays enough. It tells you what “enough” would be measured against.

Name the number

Here’s the control we’re proudest of. Pick a group, say what share of Vermont’s income tax it should carry, and the meter tells you what that means.

Say the top 5 percent, about 16,500 returns, should pay 60 percent of the income tax instead of the 51.5 percent they pay now. Holding everyone else where they are, that’s $248 million more a year. Their effective rate on AGI goes from 6.1 percent to 8.6. Scale today’s 8.75 percent top bracket in proportion and the top marginal rate lands near 12.4 percent, which would rank Vermont second in the country behind California.

Say the top 1 percent should pay half. Their effective rate goes to 15.8 percent and the implied top bracket passes 20. No state is in that neighborhood.

The meter labels all of this static arithmetic, because it is. It assumes nobody moves, retires, defers a sale or restructures income. Real yield falls as rates rise. How much is argued, not known, and we’d rather say so than bury a guess in a coefficient.

The point isn’t that any of these numbers is wrong. The point is that the slogan has never had to produce one. Now it does.

What line 11 can’t see

Every lens on the page, including the one the Legislature enacted, judges a household on a single year’s adjusted gross income. That number is precise, auditable, and blind.

It can’t tell a salary from a sale. A business sold after thirty years lands on one return as if it were an annual income, and that year’s return gets ranked in the top 1 percent next to people who earn that every year. The meter shows what the same household looks like averaged over five years. It is usually a different percentile.

It can’t see capital at risk. Money reinvested in a Vermont business isn’t deductible until the business is sold or lost. On the return, it is taxed as if the owner spent it on themselves.

It can’t see wealth, because wealth isn’t on the return. A $4 million portfolio yielding 3 percent and a $120,000 salary both show up as $120,000 of AGI. A proposal that says “wealth” but taxes “income” is taxing the second one and should be read that way.

And it can’t see hours. This is where the argument usually turns personal, so here’s what the research actually says.

A Harvard Business School diary study tracked 27 chief executives for three months and found they averaged 62.5 hours a week, worked on 79 percent of their weekend days and 70 percent of their vacation days. Labor economists have documented since the 1980s that the 50-plus-hour week grew fastest among the highest-paid, most-educated workers, reversing the old pattern in which the rich worked less than the poor.

So the executive on the phone at the lake is real. So is the other side of the same data. The bottom of the distribution is not a fixed 40 hours either. It is two jobs at 65, involuntary part-time at 28, and schedules nobody chose. Family caregivers put in somewhere around two dozen unpaid hours a week on top of whatever they’re paid for.

Both of those are true. Neither is on a tax return. And a 62-hour week versus a 40-hour one is a difference of about 1.5 times, while the average top-1-percent return in Vermont reports about 28 times the median. Hours open the conversation about effort. They don’t close it, and anyone using them to close it, in either direction, is doing the same thing the slogan does.

Three households, one number

The meter ends with three fictional Vermont families. Same filing status, same home, same $200,000 of adjusted gross income.

One inherited a business and draws dividends from it at about 20 hours a week. One built a business over 20 years, personally guaranteed its debt, and draws a salary while working the hours in that Harvard study. One is a hospital nurse in a two-earner household who also spends 24 unpaid hours a week caring for a parent.

Run them through the equal-share lens. Same answer. Proportional. Same answer. As enacted. Same answer, to the dollar, because every lens reads one line.

There is exactly one tax on the page that notices the difference between them, and it’s the federal payroll tax. It runs the other way. The founder and the nurse pay about $13,800 of it. The heir pays none.

We didn’t build the three households to tell you which one deserves what. We built them so you’d have to say which standard you’re using, and then apply it to all three.

Count the questions

One last thing the meter makes visible. Before Vermont decides what you owe, the income tax asks your filing status, your income, your dependents and their ages, your wages, your home’s value, your medical bills, and whether any of it crossed a dozen thresholds. That’s the paying side.

A universal program asks the person receiving it one question: do you live here?

Neither side of that is wrong on its face. There are good reasons to means-test contributions and good reasons not to means-test a doctor’s visit. But a “fair share” argument that stacks every condition on one side of the ledger and none on the other has made a choice, and it should say so. Right now it doesn’t, and the people making it rarely get asked.

Go play

The Fair-Share-O-Meter is live in Tax Land, next to the Spend-O-Meter and the Citizens’ Checkbook. Every constant it uses is listed on the page with its source and can be changed. Every row is tagged measured, estimated or convention. Nothing you type leaves your browser.

It won’t tell you what’s fair. We can count every dollar a Vermonter earns and every dollar the state collects. We cannot count from those figures how hard the money was earned, what anyone started with, what they risked, what they got lucky on, or what one person owes another. “Pay your fair share” sounds like arithmetic. The word fair is a moral judgment the arithmetic cannot supply on its own.

The meter shows you where the arithmetic stops. The arguing is yours.