| Illustrative use | Approximate annual amount |
|---|---|
| Cut personal income tax by about one-third and reduce corporate tax to about 19% | C$17.0-22.2B |
| Service a negotiated federal debt share at the amount currently attributed to Albertans | C$6.2B |
| Retire Budget 2026’s projected C$108.9B taxpayer-supported debt over ten years | ~C$11B |
| Increase Indigenous funding beyond continuation under a negotiated arrangement | a further C$1.5-2.8B |
| Build a sovereign wealth fund at a substantial annual contribution rate | C$10-15B |
These are illustrations of scale, not a recommended budget. Two or three may fit together; all of them do not.
Tax relief as an illustration
The analysis tests what large tax reductions would cost. Cutting personal income tax roughly in half and lowering the combined corporate rate to around 19 per cent would cost about C$24.0-31.1 billion a year. That is larger than the conservative annual balance and therefore should not be presented as an immediate promise.
| Household | Income | Income tax today | Independent Alberta | Saving | Cut |
|---|---|---|---|---|---|
| Single | $60,000 | ~$9,000 | ~$4,800 | $4,200 | 47% |
| Couple | $125,000 | ~$22,000 | ~$12,200 | $9,800 | 45% |
| Family | $165,000 | ~$26,000 | ~$13,400 | $12,600 | 48% |
| Single parent | $85,000 | ~$14,000 | ~$6,300 | $7,700 | 55% |
These examples show scale only. The fiscal case does not depend on adopting these rates.
An illustrative sequencing, and an unglamorous one:
- See what the revenue actually comes in at
- Settle the big liabilities
- Pay down debt where it makes sense
- Only then cut taxes — against money you have counted rather than money you have forecast.
Tax reduction is the most direct of these choices. The illustrative programme is not a recommendation, but the scale belongs in household terms. The savings are a permanent change in what a working household keeps, every year, for a working life.
Debt is the least visible and among the most valuable. Budget 2026 projects Alberta’s taxpayer-supported debt at C$108.9 billion for the end of 2026-27, serviced out of the budget every year before a single service is delivered. Retiring it over a decade removes that charge permanently, and a government without debt service has substantially more room to absorb a bad year without service cuts.
The Heritage Fund is a generational change. A contribution in the order of C$10 to C$15 billion a year is a different instrument from the fund as it stands: at that rate it becomes a material revenue source within a working lifetime, and the share of the budget exposed to the oil price falls every year it runs. The relevant lesson from Norway is not the size of the fund but the discipline of paying into it in the good years.
Investment in the country has the longest payback and the clearest case. Transmission, water, rail and road capacity, and the port and corridor access Alberta’s exports have never had enough of, are what raise the value of an hour of Albertan work.
None of this is in the numbers. This paper assumes no growth from lower taxes, no investment drawn in by a more predictable regime, no return on anything the Heritage Fund holds and no productivity gain from anything built.
Every one of those would improve the result, some of them substantially. They are excluded so that the case rests on what can be defended rather than on what could be hoped for.
What is out of scope in the other direction
Risks also sit outside a Day 1 fiscal ledger on the downside, and they should be named rather than left for an opponent to find. This paper does not cost continued access to the North American trade agreements on the terms Alberta enjoys today, nor a period of non-cooperation by the rest of Canada, nor the cost of standing up payment systems and a lender of last resort, nor passport issuance and international recognition, nor any gain or loss from migration after independence. Those are transition and institutional questions rather than steady-state running costs, and they are addressed in the Alberta Transition Plan and its supporting analysis rather than here.
