On August 10, 2026, Barrick Mining Corporation reported second-quarter results for the period ended June 30. The company produced 796,000 ounces of gold, above its Q2 guidance range of 730,000 to 770,000 ounces. Yet adjusted earnings per share were $0.82, below the $0.88 average estimate compiled by LSEG and reported by Reuters. For a group FP&A lead, those outcomes are not contradictory: company production guidance and an external adjusted-profit consensus are different benchmarks that govern different decisions.
Barrick also reported gold cost of sales of $1,993 per ounce and all-in sustaining costs of $1,866 per ounce, while leaving full-year production and cost guidance unchanged. Its Q2 MD&A and unaudited financial statements separately record Mali-related royalties, penalties and interest within other expense and a broader adjusted-earnings reconciliation. Barrick did not disclose a standalone amount for each component of that adjustment bundle, so those items should not be presented as the direct cause of the adjusted-EPS miss.
What changed and what it means
A favourable output variance can overstate repeatable performance when unit costs, adjusted items, jurisdiction exposure and cash conversion are not bridged separately.
- Decision affected
- Decide whether Barrick's Q2 production beat justifies raising margin and earnings assumptions, or whether higher unit costs, weaker cash conversion and separately modelled jurisdictional items require a different forecast and risk case.
- Evidence in brief
- Barrick's Q2 release and MD&A establish the guidance beat, sequential cost and cash changes, adjusted-item reconciliation and unchanged full-year guidance; Reuters attributes the $0.88 consensus to LSEG.
- What remains unresolved
- Barrick does not disaggregate the $236 million other-expense adjustment bundle by component, and the Mali tax reconciliation remains ongoing.
- Next verification
- Reconcile the Q2 volume, price, unit-cost, adjustment and cash bridges, then retest the forecast at any guidance change or the Q3 results.
Key takeaways
- Barrick produced 796,000 ounces of gold in Q2 2026, 26,000 ounces above the top of its quarterly guidance range, while adjusted EPS of $0.82 was below the $0.88 LSEG consensus reported by Reuters.
- From Q1 to Q2, gold production rose 11%, but the realised gold price fell 8%, gold AISC rose 9%, adjusted net earnings fell 17% and attributable free cash flow fell 88%.
- Lower grades, fuel and royalties explain the operating-cost pressure; Mali-related royalties, penalties and interest belong in a separate reported-to-adjusted and jurisdiction-risk bridge.
- FP&A should keep volume, unit costs, adjusted items, jurisdiction exposure and cash conversion on separate forecast lines before changing the margin outlook.
The production beat and profit miss use different benchmarks
The official Q2 release says production exceeded the company’s quarterly range and that full-year production and cost guidance remained unchanged. Reuters, using LSEG data, compares adjusted EPS with an external analyst consensus. One is a management operating range; the other is a market expectation for a non-GAAP earnings measure.
| Measure | Q2 result | Comparison | FP&A interpretation |
|---|---|---|---|
| Gold production | 796,000 ounces | Company Q2 guidance of 730,000 to 770,000 ounces | Favourable output variance against the operating range |
| Adjusted EPS | $0.82 | LSEG average estimate of $0.88 reported by Reuters | Unfavourable variance against an external earnings expectation |
| Full-year gold production | Guidance unchanged at 2.90 to 3.25 million ounces | Previously issued company guidance | No public company reforecast of the annual range |
| Full-year gold AISC | Guidance unchanged at $1,760 to $1,950 per ounce | Previously issued company guidance | Q2 cost pressure has not yet produced a public annual reset |
A single red-or-green score would erase the comparison base. FP&A should label each variance by benchmark and owner before deciding whether it changes the internal forecast, management target, external narrative or risk case. The Vestas 2026 margin bridge applies the same test to a 9.4% quarter and a raised 7–9% full-year range.
Build the volume-to-margin bridge before changing the forecast
Barrick’s first-quarter release had set the Q2 production range after reporting 719,000 ounces in Q1. The Q2 MD&A then shows that the favourable volume movement arrived alongside a lower realised gold price and higher unit costs.
| Measure | Q1 2026 | Q2 2026 | Sequential change |
|---|---|---|---|
| Gold production | 719,000 ounces | 796,000 ounces | Up 11% |
| Realised gold price | $4,823 per ounce | $4,417 per ounce | Down 8% |
| Gold cost of sales | $1,922 per ounce | $1,993 per ounce | Up 4% |
| Gold AISC | $1,708 per ounce | $1,866 per ounce | Up 9% |
| Adjusted net earnings | $1.648 billion | $1.363 billion | Down 17% |
| Attributable free cash flow | $1.213 billion | $141 million | Down 88% |
Barrick attributed the year-over-year increase in gold costs to lower grades processed at Carlin, Cortez and North Mara, higher fuel costs across operations and higher royalties associated with the stronger realised gold price. Those drivers should be modelled separately. Grade and site mix affect tonnes and recovery; fuel affects the operating-rate assumption; royalties can rise with price even when other cost controls hold. The On Holding channel-and-margin bridge extends that control to DTC mix, wholesale sell-in, price discipline, freight, currency and tariffs.
Keep operating costs and Mali-related items in separate forecast lines
The MD&A records $209 million of Q2 other expense, mainly related to additional royalties, penalties and interest arising from the retrospective application of Mali’s 2023 Mining Code to Loulo-Gounkoto for 2024 and 2025, together with remobilisation costs. A positive revaluation of Hemlo contingent consideration partly offset those amounts.
The adjusted-net-earnings reconciliation uses a different figure: $236 million of other-expense adjustments. Barrick says that bundle also includes a fair-value increment on inventory after it regained control of Loulo-Gounkoto, North American IPO legal and consulting costs, Mali remobilisation costs and the Hemlo revaluation. The components are not individually quantified in the reconciliation.
That distinction controls the causal language. Barrick adds the $236 million bundle back when moving from reported net earnings to adjusted net earnings. The LSEG comparison is also an adjusted-EPS comparison. Higher operating costs and the lower sequential realised gold price are therefore relevant to the adjusted result, while the Mali-related items are a separate reported-to-adjusted and jurisdiction-risk issue.
Barrick also disclosed that $200 million was paid to Mali in April under the tax reconciliation process and that approximately $48 million of additional penalties remained owing after a July 1 notification. Those cash and obligation disclosures should not be treated as interchangeable with the $209 million Q2 other expense or the $236 million adjustment bundle. The company said engagement with Mali on the tax reconciliation process was continuing.
Cash conversion needs its own variance line
Production and adjusted profit do not answer the cash question. Attributable operating cash flow fell from $1.968 billion in Q1 to $1.119 billion in Q2, while total attributable capital expenditure rose from $755 million to $978 million. Barrick’s attributable free cash flow therefore moved from $1.213 billion to $141 million.
The arithmetic is visible: attributable operating cash flow less total attributable capital expenditure equals attributable free cash flow for both quarters. FP&A should preserve that bridge rather than attributing the cash movement to production, costs or Mali alone. Barrick classifies attributable operating cash flow, attributable free cash flow, AISC and adjusted net earnings as non-GAAP measures without standardised meanings under IFRS, so the model should retain the company’s definitions and reconciliation. JLR’s Q1 free-cash-flow decomposition extends that control to cash profit, product investment and working-capital timing.
Unchanged full-year guidance is a constraint, not a verdict
Barrick maintained 2026 gold production guidance of 2.90 to 3.25 million ounces and gold AISC guidance of $1,760 to $1,950 per ounce. Q2 AISC of $1,866 sits inside that annual range, but the ranges cover the full year and do not establish the expected landing point after the second quarter.
The public guidance should remain one controlled reference. The internal forecast should remain the latest expected outcome, with the target-and-expectation boundary recorded explicitly. A production beat can improve the volume case without authorising FP&A to reset the margin case, cash case or jurisdiction-risk scenario. The Applied Materials Q3 forecast-state hierarchy extends that control by separating a formal quarterly guide from calendar-year expectations and customer visibility extending to 2030.
What FP&A should lock for the next forecast
- Benchmark register: label company guidance, management targets, prior forecast, prior quarter and external consensus separately.
- Operating bridge: assign volume, realised price, grade, mix, fuel, royalties and sustaining-cost movements to named owners.
- Reported-to-adjusted bridge: retain Barrick’s reconciliation and do not allocate the $236 million bundle to individual causes without disclosed support.
- Jurisdiction scenario: track paid amounts, remaining obligations, cash timing and the next Mali reconciliation disclosure outside the normal operating run rate.
- Cash bridge: reconcile attributable operating cash flow, attributable capital expenditure and attributable free cash flow before changing liquidity or capital assumptions.
- Forecast treatment: use the actuals-to-forecast control cycle to decide whether each variance changes the run rate, rephases timing, activates a scenario or leaves the full-year case unchanged.
The quarter supports a favourable production conclusion, an unfavourable unit-cost and external-consensus conclusion, and a separate jurisdiction-risk conclusion. Combining them into one performance label would make the next forecast less explainable, not more accurate.