Three Hundred Jobs

Commentary on "Meta to build $13B data centre north of Edmonton, its first in Canada" · The Canadian Press

The number that will define this industry's politics is not a megawatt figure. It is three hundred.

Meta has announced its first Canadian data centre in Sturgeon County, Alberta: more than $13 billion, one gigawatt, roughly 270,000 square metres across 1,750 acres, the company's largest facility outside the United States. Three thousand construction workers at peak. Three hundred permanent operational jobs. Sixty million dollars for local roads and water. Approximately $250 million a year to the province in royalties, taxes, levies and fees.

Thirteen billion dollars. Three hundred jobs. That is roughly $43 million of capital per permanent position.

The industry has spent two years trying not to say that out loud. It is time to stop.

The critics have the better of this argument

An organiser in Vancouver called these facilities dead weight to a city: fewer jobs than the same land would support as offices, less foot traffic for local businesses, pressure on the grid. Manitoba turned down a project on the explicit basis that the employment did not justify the resources. Speakers at Vancouver council asked for housing and durable work rather than construction phases.

On the employment question specifically, they are right, and every evasion the sector offers makes it worse.

Construction jobs are real and they are temporary. Ecosystem effects are real and they are unquantified. The claim that a facility anchors a technology cluster is sometimes true and mostly asserted. None of these answers survives contact with a resident who can see a very large building consuming very large amounts of a shared resource and employing fewer people than a mid-sized grocery store.

The honest position is that data centres are not a jobs programme and should never have been sold as one. They are capital infrastructure, like a transmission line or a port. Nobody defends a substation on headcount.

Which means the benefit case has to be built somewhere else

If employment is not the return, something else has to be, and it has to be specific enough to appear in a document.

Alberta's answer is fiscal and it is the clearest one anybody has produced: roughly $250 million annually in provincial revenue, $60 million in local infrastructure, and a stated reduction of up to six per cent in the transmission portion of consumer electricity bills through the associated generation project. Whether that last figure holds is testable, and it should be tested. But it is a benefit denominated in something a household can check on a bill, which is more than most projects have offered.

Compare the alternatives. Telus told Vancouver its facilities would recover enough waste heat to warm the equivalent of 150,000 homes. That is a larger physical benefit than anything in the Alberta package, and it was presented as an environmental credential rather than as a reduction in anyone's heating cost. Nobody could point to what they got. Manitoba was offered jobs, priced them, and declined.

Three provinces, three benefit propositions, and only the one denominated in money on a bill has survived contact with the public.

What a benefits case should contain

The sector should stop resisting this and start writing it down. A defensible package has four parts, and none of them is a job count.

Fiscal contribution, stated annually and net of any concessions received. Full cost of interconnection, including the grid assets a facility requires, paid by the facility. A measurable local benefit that a resident can verify without taking anyone's word for it. And a durable commitment, because a twenty-year asset should carry a twenty-year obligation rather than a construction-phase press release.

That is a higher bar than most projects currently clear, and the operators who complain about it will be the ones still explaining themselves at council meetings in 2030.

What this means going forward

The three hundred figure is not going away, and it should not. It is an accurate description of what this infrastructure is: enormously capital-intensive, minimally labour-intensive, and permanent.

The industry's mistake was treating that as a fact to be managed rather than a fact to be built around. A port does not apologise for its headcount. It demonstrates what moves through it. This sector has not yet learned to do the equivalent, and until it does, every community that asks what it gets will be given a number that argues against the project.

VOLTEDGE

Reference: "Meta to build $13B data centre north of Edmonton, its first in Canada" · The Canadian Press · read the article