I have been some time among them now, and I can report that the species in the cold country has perfected a thing the warmer nations only attempt. It has built a government in three storeys, assigned the heaviest work to the floor nearest the street, and arranged for the rent to be collected — quietly, and in the largest part — from the floors above, where the tenant cannot see the meter.
They are very proud of the arrangement. They call it federalism, and they teach it to their young as a triumph of balance.
The citizen, who is the point of the whole apparatus, believes he pays his taxes once a year, in the spring, with a great deal of grumbling and a single form. This is the charming part. He pays them, in truth, perhaps forty times a week — each time he fills his tank, buys his bread, pours a drink, or simply continues to own the roof over his head, which the locals have decided is a taxable act of provocation renewed every twelve months.
He does not feel forty taxes. He feels one. A system that wished to be honest with him would have arranged things otherwise. This one wished to be durable, which is a different ambition entirely, and it has succeeded — for the floor nearest the street, the one that fills his potholes and pipes his water, has been given a single instrument with which to raise money, and forbidden, by the floors above, from acquiring any other.
I asked one of them why the floor doing the most pressing work was given the least means to fund it. He explained that the lower floor is not, in the legal sense, a real thing — that it exists only as a creature permitted by the middle floor, and may be granted or denied instruments at the middle floor's pleasure. He said this with no embarrassment whatever. I wrote it down twice, to be sure I had it.
The clerks who keep the books have, I am told, prepared the figures. I shall step aside and let them speak, for they have the receipts, and I have only the astonishment.
Ask a Canadian what they pay in tax and they name their income-tax bracket. The bracket is one floor of a building they have never seen the blueprint for. The structural fact beneath every figure in this dossier is that the level of government closest to the citizen has the least power to raise money and the fastest-growing list of things it must pay for.
This is not an opinion. It is the architecture, and the architecture has been documented by the people inside it for two decades. Five facts carry the argument. Each is sourced. None is softened.
Read the five together and the squeeze is plain. The lower floor owns the work and not the wallet. The gap is closed the only way a body with one revenue tool can close it: by deferring maintenance, raising the one tax it controls, and going — in the language of the practitioners — cap-in-hand to the floors above. Deferred maintenance compounds. A property tax raised to cover it is the most regressive major tax in the system, because it tracks the value of a roof, not the income beneath it.
The lower floor owns the work and not the wallet. Everything else in this dossier is that sentence, priced.
The cleanest way to expose a structural problem is to hold the human constant and move the ground beneath them. Here is the same median earner — $65,000, identical spending, identical habits — in three cities.
The Calgary earner pays the least — but not because Alberta's income tax is lowest. At this income, British Columbia's provincial income tax is competitive. Alberta wins on the total because it levies no provincial sales tax at all. The advantage is concentrated in consumption, where it is invisible, rather than on the pay stub, where it would be felt. The Albertan does not feel lucky every time he buys something. He simply is.
Widen the lens to all four households and read the effective rate — every dollar taken, as a share of income.
Two things in this picture should trouble anyone who believes the system is sorted by ability to pay. First, the low-income worker pays a higher effective rate in Vancouver and Toronto than the income-tax structure alone would predict — because sales tax is regressive, and a person who spends nearly all of what they earn is taxed on nearly all of what they earn. Second, the gradient is geographic, not fiscal. The high earner's rate climbs from 35% in Calgary to 42% in Toronto — a seven-point swing applied to a citizen whose income, skills, and contribution are identical. The tax is a function of a line on a map.
The squeeze is not an abstraction. It is a week.
That last card is the quiet alarm of the whole dossier. The federal benefit system pays the retiree the same amount in every city. The provincial and municipal stack decides how much survives. Carry that identical senior to Calgary and the margin is +$5,614. In Vancouver, +$4,530. In Toronto, +$954. Ottawa sends the same cheque. The lower floors decide whether it lands.
| City | Household | Income | Income tax | Payroll | Sales | Fuel | Property | Paid | Received | Net | Eff. |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Calgary | Low-income | 35,000 | 3,656 | 2,445 | 1,200 | 207 | 900 | 8,408 | 975 | −7,433 | 24% |
| Calgary | Median | 65,000 | 10,753 | 4,719 | 1,700 | 322 | 2,730 | 20,224 | 0 | −20,224 | 31% |
| Calgary | High | 150,000 | 38,490 | 5,770 | 3,100 | 414 | 4,680 | 52,454 | 0 | −52,454 | 35% |
| Calgary | Retiree | 32,000 | 2,996 | 0 | 1,100 | 161 | 3,380 | 7,637 | 13,251 | +5,614 | 24% |
| Vancouver | Low-income | 35,000 | 3,759 | 2,445 | 2,880 | 335 | 900 | 10,319 | 1,422 | −8,897 | 30% |
| Vancouver | Median | 65,000 | 10,313 | 4,719 | 4,080 | 522 | 2,184 | 21,817 | 0 | −21,817 | 34% |
| Vancouver | High | 150,000 | 38,498 | 5,770 | 7,440 | 671 | 4,200 | 56,578 | 0 | −56,578 | 38% |
| Vancouver | Retiree | 32,000 | 3,187 | 0 | 2,640 | 261 | 3,080 | 9,168 | 13,698 | +4,530 | 29% |
| Toronto | Low-income | 35,000 | 3,766 | 2,445 | 3,120 | 171 | 900 | 10,401 | 1,135 | −9,266 | 30% |
| Toronto | Median | 65,000 | 10,391 | 4,719 | 4,420 | 266 | 4,828 | 24,623 | 0 | −24,623 | 38% |
| Toronto | High | 150,000 | 40,218 | 5,770 | 4,960 | 342 | 8,875 | 63,265 | 0 | −63,265 | 42% |
| Toronto | Retiree | 32,000 | 3,194 | 0 | 2,860 | 133 | 6,390 | 12,577 | 13,531 | +954 | 39% |
Two of the three big federal transfers are paid on a strict equal-per-capita basis. Every Canadian's notional share of the Health and Social Transfers is identical. Then there is equalization — the one needs-adjusted layer, and the one that generates nearly all the heat.
In 2025–26 Ottawa moved a record ~$103.8 billion through major transfers, of which $26.2 billion was equalization. Quebec received $13.6 billion; Manitoba $4.7 billion; the Atlantic provinces their shares. British Columbia, Alberta, and Saskatchewan received nothing.7 The per-capita view makes the structure legible in a way the totals do not.
The grievance from the non-recipient provinces is more specific than "we pay and they receive." The formula assesses a province's fiscal capacity — its theoretical ability to raise revenue at national-average rates — on a three-year average lagged two full years. A province that grows its economy can find its position worsened as a consequence of its own success. Critics call this a growth disincentive welded into the machinery. Defenders note, correctly, that the principle is enshrined in s.36(2) of the Constitution Act, 1982, so that a citizen in Charlottetown can expect services comparable to one in Calgary without ruinous local rates. Both are true at once. That is precisely why the layer resists reform: it does a constitutional job through a contested arithmetic.
Here is the part almost no commentary includes, and the part that turns a complaint into a diagnosis. Almost nothing in the fiscal stack can be changed by one order of government acting alone. The constitutional plumbing dictates who must agree before any idea can move — and a fix at one level routinely detonates at another. Fit For Gov calls the work of mapping this constitutional cartography.
The result is a stable, unhappy equilibrium. The municipality — closest to the pothole, the transit shortfall, the housing file — has the weakest and most regressive revenue base. The province has the tools but is constrained by the federal formula. The federal government has the money but the least direct accountability for the services the citizen actually touches. Everyone can point at someone else. No one can act alone.
The level of government you would phone to complain about your taxes is the one with the least power to change them.
A diagnosis that stops at the problem is a complaint. The brand's discipline is binary: every instrument is either Fit — keep it, fund it, deploy it — or Forward — escalate it, replace it, move. There is no "monitor" and no "convene a working group." Those are the language of the system this dossier is built to critique.
A province grants its large cities authority to levy a 1% sales tax piggybacked on existing GST/HST collection. It diversifies cities off the regressive property base onto one that grows with the economy. Toronto has already modelled it; the lever exists the moment the province writes the law.
Federal infrastructure transfers are largely restricted to capital — the bus may be bought, but not driven. Opening a portion to operating cost is a program-rules change, not new law and not new money. It addresses the exact mismatch where cities build assets they cannot afford to run.
The clearest lesson in the data: a regressive consumption tax becomes progressive when paired with an income-tested rebate paid automatically through the tax system. Ottawa already does this with the GST credit. Any new sales or fuel levy, at any level, should be bound to a paired rebate from the first day.
End the "creatures of the province" doctrine. Give cities a defined constitutional status and a guaranteed own-source revenue field. It is the cleanest cure for the vertical squeeze — and the least achievable, because it reopens the division of powers.
Re-engineer the formula to reduce the growth disincentive — shorten the lag, revisit resource inclusion, cap the clawback on provinces that expand their base. The s.36(2) principle stays untouched; only the arithmetic changes. But the arithmetic is where every province's self-interest lives.
The clerks were as good as their word. The figures are worse than I made them sound, and I had made them sound quite bad.
I had thought the cruelty was in the size of the bill. It is not. The same citizen, with the same wage and the same habits, pays a quarter of his income in one city and forty-two hundredths in another, and the difference is decided not by anything he did but by the line on the map he happened to be standing behind. The system does not sort by what a man can bear. It sorts by where he sleeps.
And the old woman on her fixed pension — the one the upper floor sends an identical cheque to, wherever she lives — keeps five thousand of it in the western city and barely nine hundred in the eastern one, because the lower floors there reach into the envelope before she has opened it. The kindness is national. The clawback is local. They have built a generosity and then arranged, very quietly, for the floor nearest the street to undo most of it.
They know all this. I want that understood. The clerks have written it down for twenty years — the creatures-of-the-province doctrine, the structural mismatch, the eight cents on the dollar for sixty cents of the work. It is not hidden from them. It is merely hidden from the citizen, who goes on believing he pays once a year, in the spring, with a single form and a great deal of grumbling.
It is a strange place, this confederation, where the order of government most fastened to the citizen is the one most starved of the instruments the citizen needs. I record it. I do not presume to mend it. The mending, as always, has been left to the floors that benefit most from leaving it undone.
The arithmetic is the argument. Every figure resolves to a public source. The model is a transparent incidence sketch of the major-money layers — federal and provincial income tax, CPP/EI, GST, PST/HST, fuel excise, property tax, and the principal benefits — computed from 2026 statutory rates. It is not a tax return: spending and fuel volumes are representative assumptions, home values are typical-not-luxury per city, and the long-tail fees (vehicle registration, tire and bottle levies, the cannabis excise stamp, land-transfer tax, liquor-board markup) are named in the apparatus but excluded from the headline, because no household pays all of them and a precise per-person figure for each would manufacture false precision. The figures expose structure and contrast, which are robust to these simplifications.
All sources public and unclassified. Figures reflect rates published as of May 2026. This dossier is independent commentary by a civic-technology practice; it is not a government publication.