BHP, the world's largest mining company, just wrote off $2.3 billion on a fertilizer mine in Saskatchewan, and this is the third time it has missed its own cost and schedule targets for the same project.

The mine is called Jansen. It sits beneath the Canadian prairie and holds one of the world's largest deposits of potash, the mineral that gives fertilizer the potassium crops need to grow. BHP approved the full two-stage development in 2021 and 2023. It has now blown past both budgets by historic margins.

Stage 1 of Jansen was approved at roughly $5.7 billion. It is now expected to cost $8.4 billion, nearly 50 percent more. Stage 2 was approved at $4.9 billion in 2023 and is now projected at $6.9 billion. First production from Stage 1 is expected in mid-2027. Stage 2, which is only 16 percent complete as of May, has been pushed to late 2031.

The original bet, and why it looked smart

BHP accelerated Jansen's development in 2022 because of the Russia-Ukraine war. Russia and Belarus together supply a large share of the world's potash. When the war began, markets feared a major fertilizer disruption, prices spiked, and the logic of building a massive Western supply alternative seemed obvious.

It turned out to be more complicated. According to BHP, the overruns were driven by inflation in materials and labor, design changes, lower productivity than forecast, and more work hours than planned to finish construction. In plain terms: the mine is harder and more expensive to build than anyone projected, and it took until now, with $2.3 billion already written off, to acknowledge how much harder.

What it means beyond one mining company's balance sheet

Potash is not a commodity most people think about, but it sits near the beginning of the food supply chain. It fertilizes the fields that produce corn, wheat, soybeans, and most other staple crops. When potash is scarce or expensive, farmers pay more. When farmers pay more, food prices feel it, eventually.

BHP's Jansen, once fully operational, was projected to supply roughly 10 percent of total global potash production. That is a meaningful share. Delays do not remove it from the market forever, but they push it further into the future at a moment when Western governments have been actively trying to reduce dependence on Russian and Belarusian supply. Stage 1 alone is not expected online until mid-2027, and the combined output of both stages is years away from a full ramp-up.

For now, the existing global suppliers, including Nutrien in Canada and state-backed producers in Belarus and Russia, retain their leverage over the market.

The deeper problem

Three cost blowouts at the same project suggest something more structural than bad luck. Large-scale mining and energy projects have a well-documented pattern of underestimating complexity, particularly underground projects where geology can surprise even experienced engineers. Optimism bias in capital approvals is real: the number that gets approved tends to be the number that makes a project viable on a spreadsheet, not necessarily the number that reflects what it costs to actually build.

BHP maintained its overall capital spending forecast at $11 billion for its 2027 financial year, meaning it is absorbing the Jansen overrun rather than pulling back elsewhere. That is a signal of continued commitment to the project, not an exit. But investors and agricultural markets watching Jansen have now learned to treat BHP's cost and timeline estimates as a floor, not a ceiling.

The impairment charge of $2.3 billion will hit BHP's earnings. The delay will hit anyone counting on a more competitive potash market before the end of the decade.