Capital Expenditure in Gold Mining: What Actually Breaks Down in Cost Control (And How to Fix It)
There is a particular kind of silence that falls over a project meeting when someone puts up a cost-to-complete slide and the numbers don't reconcile with what the GL shows.
Engineers look at the accountants. Accountants look at the engineers. The project manager looks at the ceiling. And somewhere in the background, there's a budget variance that has been quietly building for six months — because the systems weren't talking to each other, the commitment register wasn't being maintained, and nobody agreed on whether that $2.4M drill pad was capital or maintenance.
This is the reality of capital expenditure management in a gold mining operation. Not the textbook version, which involves clean feasibility studies, linear approval gates, and project costs that land exactly where the NPV model said they would. The real version involves scope changes at 11pm, contracts that span two financial years, operators charging time to the wrong WBS element, and a Finance Controller who wants a cost-to-complete in his inbox by 8am.
This article is written from the accounting seat — not the project management seat, not the treasury desk. It's about what actually goes wrong in CapEx cost control in gold mining, why it goes wrong, and what a finance team can do to bring genuine insight to the people making $50M capital allocation decisions.
First: What Makes Mining CapEx Uniquely Difficult
Before we get into the breakdowns, it's worth understanding why capital expenditure in gold mining is harder to control than in most industries.
Projects are remote, fast-moving, and physically complex. A decline extension at an underground mine might involve dozens of contractors, multiple cost centres, shared equipment, and environmental obligations — all running simultaneously in a location four hours from the nearest city. The people incurring costs are not sitting next to the people recording them.
The boundary between capital and operating expenditure is genuinely ambiguous. Is that rock bolt installation maintenance or a capital improvement? Does reprocessing old tailings capitalise? When a processing plant upgrade is done in stages across two financial years, where does capital end and repairs begin? These aren't just accounting questions — they affect royalties, tax, depreciation, and reported EBITDA. Getting them wrong has consequences.
Gold price volatility makes scope discipline hard. When gold is trading at $3,500/oz, the board approves ambitious capital programs. When it pulls back, projects get deferred mid-flight — but costs don't stop immediately. Demobilisation costs, contract cancellation fees, and partially completed assets create accounting headaches that weren't in the original capital approval.
Multiple ERP systems often coexist. At many mid-tier Australian gold miners, SAP S/4HANA manages the larger sites while smaller or recently acquired operations run on Pronto Xi, MYOB, or legacy systems. The consolidation challenge — reconciling project costs across platforms with different cost hierarchies — is substantial.
Depreciation is inseparable from CapEx. In mining, a poorly managed capital project doesn't just create a budget variance — it creates a depreciation problem that will affect the P&L for the life of the asset. Incorrect componentisation, wrong useful life assumptions, or assets that sit in CIP (Capital Work in Progress) for too long all have lasting financial consequences.
The Seven Breakdowns: Where CapEx Cost Control Actually Fails
1. The Budget Is Set Once and Never Revisited
The most common failure mode in CapEx management isn't fraud or mismanagement. It's a budget that was set eighteen months ago based on a feasibility study, and hasn't been formally revised since — even though scope has changed, contractors have re-priced, and two major items have been deferred to next year.
The budget that lives in the capital approval document and the budget that's actually being managed in the field have diverged. But because no one has formally updated the approved capital, the variance reporting is meaningless. Finance reports a 12% overspend. Operations says they've actually come in under on the scope they actually executed. Both are technically correct and completely talking past each other.
The fix: A capital budget isn't a one-time event. It's a living document that requires a formal revision process every time scope materially changes. In practice this means:
A defined threshold (often 5–10% of approved project value) that triggers a formal budget revision request to the board or investment committee
A change order log maintained jointly by the project manager and the finance team, capturing every scope change, its cost impact, and whether it's within delegated authority or requires approval
The distinction in the GL between the original approved budget, approved changes, and revised budget — all three should be visible in your capital reporting
In SAP Project System (PS), this is managed through the network and WBS budget structure. The original budget is set on release of the WBS element; approved changes are posted as budget supplements or transfers. If your SAP PS implementation isn't capturing this distinction, you're reporting against a number that doesn't mean what anyone thinks it means.
2. Commitments Are Invisible Until They Become Invoices
Here is a scenario that plays out in mining finance teams constantly: A project manager signs a $4M contract with a civil contractor in September. Nothing appears in the cost reports until January, when the first progress claim arrives. By March, the project looks dramatically over budget — even though everyone in the room knew about the contract six months ago.
The culprit is a commitment register that either doesn't exist, isn't maintained, or isn't integrated with the financial reporting. Commitments — purchase orders, contract awards, signed variations — are real financial obligations the moment they're executed. Waiting for the invoice to record the cost isn't just poor accounting; it actively misleads the people making decisions.
Why this happens more in mining than other industries:
Large contracts are often awarded verbally or with a letter of intent before the formal PO is raised
Contract values change constantly as scope evolves and productivity rates vary from estimate
In multi-site operations, procurement teams at site level may be executing contracts outside the central ERP system
Contractors often invoice in arrears, sometimes 45–60 days after the work is done
The fix: The commitment register is the single most important document in CapEx cost control. It should show, for every project and cost category:
Contracted value (the full contract award)
Variations to date (approved scope changes)
Revised contract value
Invoiced to date
Remaining commitment (revised contract value minus invoiced)
Cost-to-complete estimate (which may differ from remaining commitment if the project is running ahead or behind)
In SAP, purchase orders post as commitments automatically when goods receipt hasn't occurred — this is one of the core functions of PS integrated with MM (Materials Management). But it only works if POs are actually being raised before work starts. One of the most impactful things a finance team can do is enforce the discipline that no work commences until a PO exists in the system. It sounds basic. It is relentlessly hard to maintain on a live mine site.
The commitment register, reconciled monthly to the GL and to outstanding POs, gives you a forecast rather than just a history — which is what capital management actually requires.
3. Cost-to-Complete Is Confused With Budget Remaining
These are not the same thing, and confusing them causes some of the most damaging errors in capital reporting.
Budget remaining = Approved budget minus actual spend to date. This is a backward-looking number. It tells you how much you've used. It says nothing about whether you'll finish the project within the remaining balance.
Cost-to-complete (CTC) = The best current estimate of what it will cost to complete the remaining scope. This is a forward-looking number. It requires judgment, not just arithmetic.
A project can have $3M of budget remaining and a cost-to-complete of $5M. That's a $2M problem that won't show up in the variance report until it's too late. Equally, a project can have only $500K of budget remaining but a cost-to-complete of $300K because the hard scope is done and only minor commissioning work remains — that's a project that will land under budget.
In gold mining, CTC is particularly difficult to estimate because:
Geotechnical conditions affect civil and mining costs in ways that are hard to predict until you're in the ground
Contractor productivity rates vary significantly from estimate
Commissioning costs on processing plant upgrades are notoriously uncertain
Gold price-linked decisions (defer, accelerate, modify scope) affect the denominator mid-project
The fix: The monthly capital report to senior management should include both budget remaining and CTC, with a brief narrative explaining the key drivers of any divergence between them. CTC should be owned jointly — the project manager owns the scope assumptions, the finance team owns the cost rate assumptions and the translation into dollars. Neither can do it alone.
For large projects (typically those over a defined threshold, say $5M), a formal CTC update every quarter — reviewed by both operations and finance — is best practice. For smaller capital works, a less formal but still documented estimate is appropriate.
The format matters too. A simple three-column table — Approved Budget | Forecast Total Cost | Variance — at the category level (civil, mechanical, electrical, instrumentation, commissioning, project management) gives senior management far more insight than a single-line total.
4. The Capital / Operating Boundary Is Inconsistently Applied
Few accounting judgments have more downstream consequences in mining than the capitalisation decision. Get it wrong and you've affected:
Current year EBITDA and net profit (operating costs vs. depreciation)
Asset carrying values on the balance sheet
Future depreciation charges and asset impairment assessments
State royalty calculations (in some jurisdictions, royalties are calculated on revenue less certain capital allowances)
Tax treatment (immediate deduction vs. capital allowance)
The challenge is that the capitalisation criteria under AASB 116 (Property, Plant & Equipment) and AASB 138 (Intangibles — relevant for exploration) require judgment, and that judgment is being applied by dozens of people across a mining operation who may have different understandings of the policy.
Common inconsistencies in practice:
Maintenance shutdowns where some costs are routine (expense) and some extend asset life (capitalise) — the line gets blurry and is often drawn differently by different supervisors
Underground development: which metres are brownfield exploration (likely expense) and which metres are accessing an ore body that meets the definition of a reserve (capitalise)?
Spare parts: AASB 116.8 allows certain major spare parts to be capitalised as PP&E if they meet the recognition criteria, but this is inconsistently applied
Rehabilitation provisions: initial recognition of the ARO (Asset Retirement Obligation) is capitalised; subsequent changes have different treatment depending on the driver
Studies and feasibility work: pre-decision costs are typically expensed; costs incurred after a decision to develop are capitalised — but the decision point is not always clearly documented
The fix: A written capitalisation policy is not optional. It should define, in plain terms that a site accountant and an operations supervisor can both understand:
The capitalisation threshold (dollar amount below which items are expensed regardless of nature, typically in the range of $1,000–$5,000 depending on company size)
Category-specific guidance for the asset types common to your operation (underground development, surface infrastructure, processing plant, mobile fleet, information systems)
The documentation required to support a capitalisation decision
Who has authority to make the call for amounts above a defined threshold
The policy should be reviewed annually and communicated actively — not just filed. In practice, the most effective tool is a one-page decision flowchart that site accountants can apply in real time.
5. WBS Structure Doesn't Reflect How Management Thinks About the Project
This is a deeply practical problem that sits at the intersection of project management and accounting systems, and it causes an enormous amount of unnecessary reconciliation work.
The WBS (Work Breakdown Structure) in SAP PS — or the equivalent project hierarchy in whatever ERP your site uses — determines how costs are collected, how budgets are tracked, and how reports are produced. If that structure doesn't map to how the project manager thinks about the work, and how the CFO thinks about the project, you will spend an inordinate amount of time manually re-cutting reports in Excel.
Common failure patterns:
WBS is structured by accounting category (civil costs, mechanical costs, labour) but management wants to see costs by scope area (shaft development, headframe, winder installation)
WBS is structured by cost type but contracts span multiple cost types — so a single contractor's costs are split across five WBS elements and the commitment tracking becomes impossible
WBS hierarchy is too flat (one level, no sub-elements) so there's no ability to report at a granular level without manual analysis
WBS hierarchy is too deep (five or six levels) which makes data entry burdensome and leads to people posting to wrong elements
The fix: The WBS structure should be designed before the project starts, in a joint session between the project manager, the finance team, and (if using SAP) the SAP PS administrator. The design should answer three questions:
What is the level at which costs need to be controlled? (This determines the lowest-level WBS element that carries a budget)
What is the level at which management wants to see reporting? (This determines the roll-up hierarchy)
How do the main contracts and purchase orders map to WBS elements? (This determines whether commitment tracking will be practical)
In SAP PS, a well-designed WBS also enables automated settlement to the fixed asset (via the capital project settlement rule), which feeds directly into Asset Accounting (FI-AA) and ensures the depreciation start date and asset class are correctly populated when the asset is capitalised. A poorly designed WBS makes this settlement process manual and error-prone.
6. The Monthly Capital Report Is Historical, Not Insightful
Walk into most mining company finance teams and ask to see the monthly capital report. You will usually see:
Actuals vs. budget, current month
Actuals vs. budget, year to date
Budget remaining
That's it. It's a rearview mirror. It tells you where you've been. It doesn't help a CFO or General Manager make a decision.
The information that would actually be useful for decision-making — and which is entirely producible with the data already in the system — is almost never in the standard report.
What good capital reporting actually includes:
Forecast total project cost vs. approved budget. Not just spend to date, but the full picture of where this project will end up. This requires a CTC (see point 3 above) but once you have it, this is the single most decision-relevant number in capital reporting.
Commitment analysis. Contracted-but-not-yet-invoiced amounts, broken down by major contract. This tells management what's already locked in and what flexibility remains.
Schedule vs. cost performance indicators. Is the project spending fast because it's ahead of schedule (good) or because costs are running above estimate (bad)? These look identical in a pure cost report. Basic earned value concepts — even informal ones — help distinguish the two.
Key risks and the financial exposure. Every project has two or three items that are genuinely uncertain. The capital report should name them and quantify the range. "Groundwater management costs: base case $800K, upside $500K, downside $1.5M" tells management something they can act on.
Look-back on completed projects. Monthly, include one project that has been formally closed out in the past quarter. How did final cost compare to approved budget? What were the drivers of variance? This builds organisational learning and, over time, improves budget accuracy.
Capitalisation status. What is sitting in CIP? What is expected to be capitalised in the next quarter? Are there any assets that have been in CIP for more than twelve months and need a status review? (Prolonged CIP balances are a common audit focus area.)
The format should be designed for its audience. An executive capital report should fit on two pages with a clear narrative. The supporting detail — commitment register, WBS-level actuals, contractor progress claims — sits behind it for those who need to drill in.
7. Project Close-Out Is Treated as an Accounting Task, Not a Business Event
When a capital project is operationally complete, a series of accounting events need to happen: the CIP balance needs to be settled to the fixed asset register, the asset needs to be componentised correctly, a useful life needs to be assigned, depreciation needs to commence, and the WBS element needs to be technically closed to prevent further postings.
In many organisations, this process is treated as an administrative accounting task — something the finance team does when they get around to it. The result is a fixed asset register full of CIP balances that should have been capitalised months ago, depreciation that is understated, and assets that are generating no depreciation expense despite being in full productive use.
This isn't just a balance sheet issue. It's a profitability measurement issue. If a $20M processing plant upgrade has been commissioned but not capitalised, you're understating depreciation — and therefore overstating profit — on your operational assets.
The fix: Project close-out should be a defined process with a checklist, clear ownership, and a target timeframe from operational completion to full capitalisation. In practice, sixty days from commissioning to capitalised asset is a reasonable target for most projects; larger or more complex projects might warrant ninety days.
The close-out checklist should include:
Final cost reconciliation: total actual cost vs. approved budget, with variance explanation
Commitment clearance: confirmation that all POs are closed and all invoices received (or accrued)
Componentisation schedule: how the total project cost is split across asset components, each with their own useful life
Asset master data: location, cost centre, asset class, useful life — reviewed and confirmed before posting
Settlement to fixed asset register
WBS technical completion in SAP PS
Post-implementation review: was the project delivered on scope, on budget, on schedule?
The post-implementation review is the step most commonly omitted and most valuable in the long run. It closes the loop between the investment case (what we said this project would do) and the outcome (what it actually did). Done consistently, it builds the data that improves future capital decisions.
Bringing It Together: The Capital Accountant's Role in Strategic Decision-Making
The seven breakdowns above are operational problems, but fixing them serves a strategic purpose: giving senior management the information they need to allocate capital well.
Gold mining is, at its core, a capital allocation business. The decisions that determine whether a company creates or destroys value — which projects to fund, which to defer, where to apply sustaining capital, when to invest in exploration vs. processing optimization — are capital decisions. The quality of those decisions depends entirely on the quality of the financial information underlying them.
A finance team that produces timely, accurate, forward-looking capital reports isn't doing administration. It's providing strategic infrastructure.
The specific contributions a strong capital accounting function makes to strategic decision-making:
Portfolio view, not project-by-project. At any given time, a mid-tier gold miner might have fifteen to twenty active capital projects ranging from $200K equipment purchases to $50M expansion projects. A portfolio view — total committed capital, forecast total spend, capital at risk — enables the CFO to see whether the company is within its capital allocation envelope and where the risks are concentrated.
Early warning on overruns. A robust commitment register and CTC process surfaces overrun risk before it becomes an overrun fact. This gives management options — defer a scope element, accelerate a contractor, revisit the investment case — that disappear once the money is spent.
Input to capital allocation decisions. When a new project is being evaluated against the existing capital program, finance needs to be able to answer: Do we have the available capital? Which existing projects could be deferred if necessary? What's the opportunity cost? This requires a dynamic view of the capital portfolio, not just the current year budget.
Asset register quality for impairment assessments. The carrying values of mining assets are scrutinised every year for impairment indicators. Accurate capitalisation, correct useful lives, and properly maintained asset registers are the foundation of a defensible impairment assessment.
Depreciation forecasting. The capital decisions made today determine the depreciation profile for the next ten to twenty years. A finance team that can model the depreciation impact of different capital programs — accelerated investment now versus a more conservative program — is contributing directly to forward-looking financial planning.
A Note on ERP Systems and the Gap Between Capability and Reality
SAP S/4HANA has the functionality to manage all of the processes described in this article elegantly. Commitment management, budget version control, WBS hierarchies, settlement to fixed assets, asset componentisation — it's all there.
The gap, in almost every organisation that uses it, is between what the system can do and what the people using it actually do.
POs that don't get raised until after the invoice arrives. WBS elements that get created at the last minute with no thought to structure. Assets that settle to the wrong category because the settlement rule was never configured properly. Budget updates that happen in a spreadsheet rather than in the system.
The technical capability of the ERP is a ceiling, not a floor. The actual quality of your CapEx data is determined by the discipline of the processes around it — the controls, the training, the month-end routines, and the willingness of the finance team to push back when the operational teams want to bypass the process.
That advocacy role — insisting on the process, explaining why it matters, and translating the accounting requirements into terms that make sense for people whose primary job is building mines — is one of the most valuable things a capital accountant does.
Conclusion
Capital expenditure management in gold mining is hard not because the accounting is technically complex — though it can be — but because the environment is genuinely difficult: remote, fast-moving, politically charged, and subject to external forces (gold price, ground conditions, labour availability) that no budget model fully captures.
The response to that difficulty isn't to accept poor data and produce reports that no one trusts. It's to build the processes, disciplines, and relationships that produce information worth having — and then use that information to help the people running the business make better decisions.
The commitment register that gets maintained. The WBS structure that was designed before the project started. The CTC conversation that happens monthly, not quarterly. The close-out review that actually captures what was learned. The capital report that tells a story rather than just displaying numbers.
None of these are heroic acts. They're the craft of the job — done consistently, in a complex environment, by finance professionals who understand both the accounting and the mining operation they're accounting for.
That combination — technical rigour and operational understanding — is what good capital accounting looks like.
Comments