Alberta’s separation debate has entered a new phase. Elections Alberta recently declared the independence petition successful, with 222,597 verified signatures, above the 177,732 required. Further steps await Alberta’s Court of Appeal review of a lower court decision that invalidated the petition’s approval. Separately, Alberta has scheduled a non-binding referendum for Oct. 19 on remaining in Canada or starting the constitutional process toward a binding separation vote.Much of the debate focuses on how much money Albertans send to the federal government. A 2026 Library of Parliament study indicates that Alberta is a net contributor, having made the largest provincial net contribution in 2024. But that figure captures only a part of the economic picture. Less obvious aspects may be more important.My working paper constructs a broader balance sheet. It estimates gains of $1,075 per resident from taxes and pension against losses of $6,996 in obligations. In other words, negotiated independence would cost more than it saves. Nonetheless, these results should be read with caution. They are estimates based on reasonable assumptions about future economic scenarios, not statistical forecasts.More than a stay-or-leave choiceSecession, the formal withdrawal of a region from a country, involves trade-offs. A larger population provides broader markets and shares the costs and risks of public programs. In exchange, independence provides control and policy more responsive to local preferences.In Canada, separation would require negotiation. The Supreme Court of Canada’s Québec secession reference says a majority in a referendum would create an obligation to negotiate, not an automatic right to leave. The Clarity Act says lawful secession requires a constitutional amendment and negotiations over assets, liabilities, borders, Indigenous rights and minority rights.I compare four institutional scenarios for Alberta: The status quoInside-Canada autonomy, including provincial tax collection, a provincial pension plan and selected new agenciesSovereignty-association, a political split where Alberta would become a country but keep a close economic partnership (like a shared currency and free trade) with CanadaNegotiated independence, in which diplomacy between Canada and Albert would arrange the peaceful transfer of power.My calculations began with observed 2023 Statistics Canada income data, interprovincial trade data and federal public-finance records.The alternative scenarios are hypothetical, defined by assumptions. For example, I assume a six per cent increase in trade and regulatory costs under independence, with 30 per cent passed on to consumers. These inputs produce an estimated 1.8 per cent increase in consumer prices. Then, I repeat the calculation using lower and higher trade cost numbers to see how the results change.Each scenario affects the Alberta economy in seven ways: 1. Additional trade and regulatory costs with the rest of Canada2. Replacing major federal transfers3. Preserving selected household benefits4. Creating new administration, with costs tied to provincial public services GDP5. Servicing a negotiated share of federal debt6. Receiving a possible fiscal dividend7. Receiving a possible pension dividendThe fiscal bill dominatesI combined these seven areas of impact to estimate purchasing power under each of the four scenarios. Compared with the status quo, average purchasing power falls by one per cent under “inside Canada autonomy,” 4.4 per cent under “sovereignty-association” and 9.3 per cent under “negotiated independence.” A second, inequality-sensitive measure considers how these gains and losses are distributed. People who currently pay more tax will bear a larger share of the costs of separation. Under independence, for instance, purchasing power falls by about 1.9 per cent for the lowest income group and 14.9 per cent for the highest income group.Another way to assess separation is the break-even test. It asks how much additional benefit each of the four scenarios would require to match the status quo.The annual amount is $252 per resident for autonomy, $1,447 for sovereignty association and $3,265 for independence. These amounts don’t measure what Albertans would be willing to pay for self-government; they represent benefits that would be paid to Albertans to leave them as well off as they are under the status quo.My study credits independence with a fiscal dividend of $850 per resident and a pension dividend of $225. This estimate draws on research into Alberta’s broader federal fiscal balance, not equalization alone. The pension estimate is deliberately modest and follows the Chief Actuary’s interpretation of Canada Pension Plan rules. Those gains do not cover the $6,996 in gross new obligations. Roughly 74 per cent comes from replacing major transfers and preserving selected working age and family benefits. New administration and debt service add to the bill. The result is not driven by an assumed economic collapse. Market income is estimated to fall by only 0.3 per cent and consumer prices to rise by only 1.8 per cent. The price estimate allows for regulatory, tax, licensing and administrative costs that can arise at a new border even without tariffs, as research on the Canada-United States border effect demonstrates.What the numbers do not tellThis calculation leaves out potentially important factors. It does not consider capital flight, currency choices, one-time transition costs, long-run productivity changes or the full legal and treaty issues raised by secession. Negotiated arrangements could reduce some costs; uncertainty could increase others.Economics cannot obviously determine the value of identity, self-government or constitutional control. The evidence does not tell Albertans how to vote. But it does show that autonomy and independence are economically distinct choices. An informed referendum debate requires a complete institutional package and a credible financing plan, not just a federal balance number.Constantin Colonescu does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.