Equal
…
…
ROE = tax × interest × margin × turnover × leverage
| FY2021 | FY2022 | FY2023 | FY2024 | FY2025 | |
|---|---|---|---|---|---|
| Tax burden | 0.72x | 0.79x | 0.44x | 0.71x | 0.87x |
| Interest burden | 0.79x | 0.91x | 0.75x | 0.76x | 0.43x |
| Operating margin | 42.9% | 54.0% | 37.7% | 44.5% | 36.5% |
| Asset turnoverdriver | 0.25x | 0.44x | 0.22x | 0.24x | 0.13x |
| Leverage (equity mult.) | 2.09x | 1.80x | 1.96x | 2.12x | 2.55x |
| = Return on Equity (consolidated) | 12.7% | 30.3% | 5.4% | 12.1% | 4.6% |
| Return on Invested Capital (ROIC) | 10.6% | 26.0% | 5.1% | 9.5% | 5.2% |
Consolidated (pre-minority-interest) basis; the headline ROE in the ratio grid is owners’ basis.
| FY2021 | FY2022 | FY2023 | FY2024 | FY2025 | |
|---|---|---|---|---|---|
| Current Ratio(Current Assets / Current Liabilities) | 2.53x | 3.45x | 3.29x | 2.02x | 2.24x |
| Quick Ratio((Current Assets − Inventory) / Current Liabilities) | 1.53x | 2.59x | 2.81x | 1.35x | 1.49x |
| Cash Ratio(Cash / Current Liabilities) | 1.02x | 1.50x | 1.59x | 0.65x | 0.48x |
| Working Capital(Current Assets − Current Liabilities) | $ 835 M | $ 1.3 B | $ 1.8 B | $ 1.2 B | $ 1.7 B |
| FY2021 | FY2022 | FY2023 | FY2024 | FY2025 | |
|---|---|---|---|---|---|
| Debt to Equity (DER)(Total Debt / Total Equity) | 0.75x | 0.51x | 0.70x | 0.82x | 1.19x |
| Liabilities to Equity(Total Liabilities / Total Equity) | 1.09x | 0.80x | 0.96x | 1.12x | 1.55x |
| Debt to Assets(Total Debt / Total Assets) | 0.36x | 0.28x | 0.36x | 0.39x | 0.47x |
| Net Debt(Total Debt − Cash) | $ 1.3 B | $ 1.0 B | $ 2.0 B | $ 3.6 B | $ 5.8 B |
| Interest Coverage(EBIT / Interest Expense) | 4.67x | 10.76x | 3.94x | 4.18x | 1.74x |
| Equity Multiplier (Assets ÷ Equity) | 2.09x | 1.80x | 1.96x | 2.12x | 2.55x |
| FY2021 | FY2022 | FY2023 | FY2024 | FY2025 | |
|---|---|---|---|---|---|
| Gross Margin(Gross Profit / Revenue) | 54.9% | 61.0% | 52.1% | 59.3% | 47.5% |
| Operating Margin(EBIT / Revenue) | 42.9% | 54.0% | 37.7% | 44.5% | 36.5% |
| Net Margin(Net Income / Revenue) | 24.4% | 38.6% | 12.4% | 23.9% | 13.5% |
| EBITDA(EBIT + D&A) | $ 687 M | $ 1.7 B | $ 944 M | $ 1.4 B | $ 997 M |
| EBITDA Margin(EBITDA / Revenue) | 52.8% | 59.4% | 46.4% | 53.7% | 54.0% |
| Return on Assets (ROA)(Net Income / Total Assets) | 6.1% | 16.8% | 2.8% | 5.7% | 1.8% |
| Return on Equity (ROE)(Net Income (Owners) / Equity (Owners)) | 13.1% | 31.0% | 5.5% | 12.3% | 4.7% |
| Tax Burden (Net ÷ Pretax) | 0.72x | 0.79x | 0.44x | 0.71x | 0.87x |
| Interest Burden (Pretax ÷ EBIT) | 0.79x | 0.91x | 0.75x | 0.76x | 0.43x |
| Return on Invested Capital (ROIC) | 10.6% | 26.0% | 5.1% | 9.5% | 5.2% |
| Turnover | FY2021 | FY2022 | FY2023 | FY2024 | FY2025 |
|---|---|---|---|---|---|
| Asset Turnover(Revenue / Total Assets) | 0.25x | 0.44x | 0.22x | 0.24x | 0.13x |
| Inventory Turnover(COGS / Inventory) | 1.07x | 2.35x | 2.66x | 1.41x | 0.93x |
| Receivables Turnover(Revenue / Receivables) | 8.43x | 8.33x | 5.14x | 9.81x | 3.02x |
| Payables Turnover(COGS / Payables) | 5.19x | 6.53x | 4.28x | 4.07x | 3.60x |
| Conversion Period | FY2021 | FY2022 | FY2023 | FY2024 | FY2025 |
|---|---|---|---|---|---|
| Days Inventory Outstanding (DIO)(365 × Inventory / COGS) | 339.7 days | 155.4 days | 137.0 days | 259.4 days | 393.7 days |
| Days Sales Outstanding (DSO)(365 × Receivables / Revenue) | 43.3 days | 43.8 days | 71.0 days | 37.2 days | 120.7 days |
| Days Payable Outstanding (DPO)(365 × Payables / COGS) | 70.3 days | 55.9 days | 85.2 days | 89.7 days | 101.3 days |
| Cash Conversion Cycle (CCC)(DIO + DSO − DPO) | 312.7 days | 143.3 days | 122.8 days | 206.9 days | 413.1 days |
| FY2021 | FY2022 | FY2023 | FY2024 | FY2025 | |
|---|---|---|---|---|---|
| Free Cash Flow(Operating Cash Flow − Capex) | $ 139 M | $ 298 M | -$ 1.6 B | -$ 1.6 B | -$ 1.8 B |
Price Rp 4,100 · market cap Rp 297 T ($ 16 B at the cited rate; statements are filed in USD)
| Multiple | AMMN | Peer median | vs median |
|---|---|---|---|
| P/E | 66.03x | 16.25x(15/16) | +306% |
| P/B | 3.08x | 1.68x | +83% |
| P/S | 8.90x | 1.92x | +363% |
| EV/EBITDA | 22.39x | 10.31x | +117% |
| EV/EBIT | 33.13x | 14.25x | +133% |
| EV/Sales | 12.09x | 2.27x | +432% |
| FCF Yield | -11.24% | 0.01% | -106,798% |
| Dividend Yield | — | 5.27%(11/16) | — |
EV = mkt cap $ 16 B + debt $ 6.5 B − cash $ 677 M + minority interest $ 93 M = $ 22 B
not computable: negative or zero base-year FCF. Shown as-is rather than estimated.
Reverse DCF: single-stage FCF, terminal growth 2.5%, discount band 9.3–13.3% (β=1; build cited in Methodology). Not a forecast.
This model projects the operating cash the whole business generates, discounts it back at the blended cost of capital, and subtracts net debt. What remains is the equity value, stated per share. Every input below comes from the company's own audited record or from cited market data, and you can adjust each one yourself.
Mechanical DCF suppressed: on these default assumptions the modelled enterprise value falls BELOW net debt and minority interests, so the equity residual is negative. Equity cannot be worth less than nothing, so no per-share figure is published here: read it as the model saying the debt claims consume the whole enterprise at this discount rate and growth path, which is itself the signal. The components are shown below so the arithmetic stays checkable, and the sliders let you test what it would take to change the answer.
Cyclical normalization: Commodity/cyclical name: the trailing years are a sample drawn from the price cycle, not a trend. Defaults are therefore normalized, using the full-window mean margin with no cycle-position growth extrapolation. That normalized margin is itself the embedded commodity-price assumption.
Base year contains named one-off item(s): FY2022: The high-grade, high-price year: revenue +118% to USD 2,830.1 M on Phase 7 ore and a strong copper market.; FY2025: The Phase 8 low-grade start plus smelter commissioning: concentrate output down 55% year on year in Q1, revenue −31%, and USD 613.8 M absorbed into inventory and receivables.. The EBIT basis screens out most non-operating items, but read the Earnings Quality section before trusting the base margin.
| Assumption | Value | Basis |
|---|---|---|
| Rf (risk-free) | 4.66% | US 10Y Treasury, 10 Jul 2026 |
| ERP (Rm − Rf) | 6.69% | Damodaran Indonesia, Jul 2026 (Baa2, CRP 2.46%) |
| β | 0.96 → 1.23 | Metals & Mining (unlevered) relevered at own D/E 0.39 |
| Cost of equity | 12.91% | Rf + β × ERP |
| Cost of debt | 5.97% | FY2025 interest expense ÷ total debt |
| Tax rate | 27.6% | median effective rate, FY2021–FY2025 (pretax ≈ EBIT − interest) |
| WACC | 10.49% | 72% E × CoE + 28% D × Kd × (1 − t) |
| Revenue growth (yr 1, fading) | 2.5% | terminal growth from year 1, cyclical normalization: the delivered 4-yr CAGR (9.2%) reflects cycle position, not a trend |
| EBIT margin | 43.1% | full-cycle mean EBIT margin, FY2021–FY2025 (cyclical normalization: this margin is the embedded commodity-price assumption) |
| D&A / revenue | 11.8% | mean D&A/revenue, last 3 FYs |
| Capex / revenue | 50.0% | mean capex/revenue, last 3 FYs, for the explicit years; in the terminal year capex falls to replacement level (equal to depreciation, 11.8% of revenue) because holding a build-phase or pause-phase ratio in perpetuity misprices the company in whichever direction that phase points |
| ΔNWC / Δrevenue | 8.5% | median ΔNWC/Δrevenue across the seeded years (NWC = AR + inventory − AP) |
| Terminal growth | 2.5% | BI inflation-target midpoint (2.5% ± 1%) |
| Year | +1 | +2 | +3 | +4 | +5 | T∞ |
|---|---|---|---|---|---|---|
| Growth | 2.5% | 2.5% | 2.5% | 2.5% | 2.5% | 2.5% |
| Revenue | $ 1.9 B | $ 1.9 B | $ 2.0 B | $ 2.0 B | $ 2.1 B | $ 2.1 B |
| EBIT | $ 816 M | $ 836 M | $ 857 M | $ 879 M | $ 901 M | $ 923 M |
| NOPAT | $ 591 M | $ 606 M | $ 621 M | $ 636 M | $ 652 M | $ 669 M |
| + D&A | $ 224 M | $ 229 M | $ 235 M | $ 241 M | $ 247 M | $ 253 M |
| − Capex | $ 946 M | $ 970 M | $ 994 M | $ 1.0 B | $ 1.0 B | $ 253 M |
| − ΔNWC | $ 3.9 M | $ 4.0 M | $ 4.1 M | $ 4.2 M | $ 4.3 M | $ 4.4 M |
| FCFF | -$ 136 M | -$ 139 M | -$ 143 M | -$ 146 M | -$ 150 M | $ 664 M |
| PV | -$ 123 M | -$ 114 M | -$ 106 M | -$ 98 M | -$ 91 M | $ 5.0 B |
Check it yourself: revenue × margin = EBIT · NOPAT = EBIT × (1 − tax) · FCFF = NOPAT + D&A − Capex − ΔNWC · PV = FCFF ÷ (1+WACC)^yr · TV = FCFF(T∞) ÷ (WACC − g), discounted from year 5
EV = PV(explicit) -$ 532 M + PV(TV) $ 5.0 B = $ 4.5 B · TV 112% of EV · − net debt $ 5.8 B − minority $ 93 M
Model output: Rp -339/share (-108% vs price Rp 4,100)· exit-multiple check (10.3x): Rp 194
Under these assumptions the model lands 108% below today's price. The market, in other words, is paying for faster growth, a fatter margin, or a lower discount rate than the inputs here assume.
| g \ WACC | 9.5% | 10.5% | 11.5% |
|---|---|---|---|
| 2.0% | -217 | -431 | -597 |
| 2.5% | -96 | -339 | -525 |
| 3.0% | 45 | -235 | -445 |
Model output under the stated assumptions. Every input above comes from the company's own audited record or from cited market data, and every one can be adjusted. This is not a price target and not advice.
Educational analysis. Multiples pair today’s price with the latest audited fiscal year (trailing-FY convention); peer figures are the covered peer sample, not the whole market. Never a price target.
| FY2021 | FY2022 | FY2023 | FY2024 | FY2025 | |
|---|---|---|---|---|---|
| Revenue | $ 1.3 B | $ 2.8 B | $ 2.0 B | $ 2.7 B | $ 1.8 B |
| Cost of Goods Sold | $ 586 M | $ 1.1 B | $ 973 M | $ 1.1 B | $ 970 M |
| Gross Profit | $ 713 M | $ 1.7 B | $ 1.1 B | $ 1.6 B | $ 877 M |
| Operating Income (EBIT) | $ 557 M | $ 1.5 B | $ 767 M | $ 1.2 B | $ 674 M |
| Interest Expense | $ 119 M | $ 142 M | $ 195 M | $ 284 M | $ 386 M |
| Net Income | $ 317 M | $ 1.1 B | $ 252 M | $ 637 M | $ 249 M |
| Net Income Attributable to Owners | $ 317 M | $ 1.1 B | $ 252 M | $ 637 M | $ 249 M |
| Depreciation & Amortization | $ 129 M | $ 154 M | $ 177 M | $ 246 M | $ 323 M |
| FY2021 | FY2022 | FY2023 | FY2024 | FY2025 | |
|---|---|---|---|---|---|
| Cash & Equivalents | $ 558 M | $ 818 M | $ 1.2 B | $ 754 M | $ 677 M |
| Accounts Receivable | $ 154 M | $ 340 M | $ 396 M | $ 271 M | $ 611 M |
| Inventory | $ 545 M | $ 470 M | $ 365 M | $ 771 M | $ 1.0 B |
| Current Assets | $ 1.4 B | $ 1.9 B | $ 2.5 B | $ 2.3 B | $ 3.2 B |
| Total Assets | $ 5.2 B | $ 6.5 B | $ 9.1 B | $ 11 B | $ 14 B |
| Accounts Payable | $ 113 M | $ 169 M | $ 227 M | $ 267 M | $ 269 M |
| Current Liabilities | $ 546 M | $ 546 M | $ 774 M | $ 1.2 B | $ 1.4 B |
| Total Liabilities | $ 2.7 B | $ 2.9 B | $ 4.5 B | $ 5.9 B | $ 8.4 B |
| Total Interest-Bearing Debt | $ 1.9 B | $ 1.8 B | $ 3.3 B | $ 4.3 B | $ 6.5 B |
| Total Equity | $ 2.5 B | $ 3.6 B | $ 4.6 B | $ 5.2 B | $ 5.4 B |
| Equity Attributable to Owners | $ 2.4 B | $ 3.5 B | $ 4.5 B | $ 5.2 B | $ 5.3 B |
| FY2021 | FY2022 | FY2023 | FY2024 | FY2025 | |
|---|---|---|---|---|---|
| Operating Cash Flow | $ 294 M | $ 991 M | -$ 121 M | $ 152 M | -$ 475 M |
| Capital Expenditure | $ 155 M | $ 692 M | $ 1.5 B | $ 1.8 B | $ 1.4 B |
AMMN gross margin: 54.9 % (2021) → 61.0 % (2022) → 52.1 % (2023) → 59.3 % (2024) → 47.5 % (2025). OPM: 42.9 % → 54.0 % → 37.7 % → 44.5 % → 36.5 %. Net margin: 24.4 % → 38.6 % → 12.4 % → 23.9 % → 13.5 %. ROE: 13.1 % → 31.0 % → 5.5 % → 12.3 % → 4.7 %. ROA: 6.1 % → 16.8 % → 2.8 % → 5.7 % → 1.8 %. ROIC: 10.6 % → 26.0 % → 5.1 % → 9.5 % → 5.2 %. D/E: 0.75 → 0.51 → 0.70 → 0.82 → 1.19. Interest coverage: 4.7× → 10.8× → 3.9× → 4.2× → 1.7×. FCF (USD): +0.14 B → +0.30 B → −1.64 B → −1.64 B → −1.85 B. Net debt (USD): 1.3 B → 1.0 B → 2.0 B → 3.6 B → 5.8 B. Revenue (USD): 1.30 B → 2.83 B → 2.03 B → 2.66 B → 1.85 B. The asset base tripled (USD 5.2 B → 13.9 B) while revenue and returns compressed in 2025 (ROIC 5.2 %, ROE 4.7 %); a classic build-phase valley: Phase 8 low grades + the concentrate-to-cathode transition suppress near-term output while the smelter, PMR and power plant absorb cash. Interest coverage at 1.7× (2025) is the pressure gauge to watch: the deposit is elite, but the balance sheet is now doing the heavy lifting until the smelter ramps to full utilisation and Phase 8 grades normalise. The 2022 peak (ROIC 26.0 %, FCF +USD 0.30 B, EBITDA margin 59.4 %) shows the through-cycle earning power of the orebody once capex rolls off.
Batu Hijau is an open pit mined in phases. Phase 7 finished at the end of 2024 and Phase 8 began in the low-grade outer halo, so most of 2025 was spent stripping waste and processing weak ore before the higher-grade central zone is reached. Q1 2025 concentrate fell 55% year on year, copper 62% and gold 81%, with the year guided at 430,000 dmt of concentrate weighted to a second-half ramp.
EconomicsThis is the single most misread feature of an open-pit miner. The waste you remove is a cost today and the ore it uncovers is revenue in a later year, so the income statement and the mine plan run on different clocks. A phase transition therefore produces accounts that look like deterioration while the asset is actually being prepared. The honest test is not the P&L, it is whether the reserves behind the spending are real and disclosed: Phase 8 carries roughly 460 Mt and extends mine life to 2030.
Batu Hijau holds about 705 Mt of reserves at 0.36% copper and 0.28 g/t gold, with Phase 8’s own 442 Mt richer at 0.38% copper and 0.36 g/t gold. Revenue across the window went USD 1,299.1 M, 2,830.1 M, 2,033.4 M, 2,663.6 M, 1,846.5 M: swings of +118%, −28%, +31% and −31%.
EconomicsTwo variables you do not control set that line: the metal price and the grade of whatever the mine plan says you dig this year. Neither responds to management effort, which is why revenue here is even less informative about operating quality than at an ordinary price-taker like BYAN. A copper miner’s revenue swinging by a third in either direction is normal; what you judge is cost per tonne moved and whether the capital being spent buys grade later.
The copper smelter, with 900,000 tonnes a year of input capacity, was commissioned through 2025: first anode in February, first cathode in March, first refined gold in July, reaching roughly 71% smelter and 55% precious-metals-refinery utilisation by Q4. It will process Batu Hijau concentrate and, later, ore from Elang, whose feasibility study completed in 2025. Concentrate export permits were extended to bridge the ramp.
EconomicsSelling concentrate means handing the smelting and refining margin to someone else, usually abroad. Owning the smelter captures it and satisfies Indonesia’s downstreaming rules, but it converts a mining company into a mining-and-processing company with a much larger fixed-asset base. The margin arrives only when utilisation is high, which is why 71% and 55% are the two numbers to track next year, not the copper price.
Capex ran USD 154.8 M, 692.3 M, 1,519.6 M, 1,791.8 M and 1,372.4 M, which is 12%, 24%, 75%, 67% and 74% of revenue. Total borrowings went USD 1,862.6 M → 6,461.6 M, debt-to-equity 0.75 → 1.19, interest expense USD 119.3 M → 386.0 M, and interest cover 4.67× → 10.76× → 1.74×.
EconomicsHere is the combination that decides the risk: capital commitments that cannot flex, sitting on top of revenue that swings with grade. Interest alone now consumes 20.9% of revenue. Compare BYAN, which met a downturn owing nothing at all, and the contrast is not about which management is smarter but about where each company sits in its own asset cycle, and how much room the balance sheet leaves if the next year disappoints.
FY2025: net profit USD 249.0 M, operating cash flow MINUS USD 475.0 M. FY2023 was the same shape, profit USD 252.1 M against operating cash flow of minus USD 121.2 M. Free cash flow has been about minus USD 1.6 to 1.8 B in each of the last three years.
EconomicsIf you take one habit from this page, take this one. Profit and cash can point in opposite directions, and when they do the cash line is usually telling you where the business actually is. Here the gap is explainable rather than sinister: inventory rose USD 274.6 M and receivables USD 339.2 M as concentrate was stockpiled and refined metal built up during commissioning, so USD 613.8 M of cash went into working capital in a year that earned USD 249.0 M. Explainable is not the same as harmless, because that stock has to convert.
Cost structureThree costs, and only one of them is mining. Cost of revenue is USD 969.8 M against USD 1,846.5 M of sales, a 47.5% gross margin that is genuinely good for a copper miner even in a low-grade year. Then depreciation, which has risen from USD 129.4 M to USD 323.3 M as the smelter and Phase 8 assets are capitalised, now 17.5% of revenue, and this is why the EBITDA margin of 54.0% looks so much healthier than the 36.5% operating margin: the gap between them is the new asset base beginning to be charged against earnings. Then interest, USD 386.0 M, which is 20.9% of revenue and more than the depreciation. A miner whose interest bill is a fifth of revenue has moved a large part of its risk from the ore body to the balance sheet.
Cash cycleLong and getting longer, which is exactly what a commissioning year does to working capital. The cash-conversion cycle ran 312.7, 143.3, 122.8, 206.9 and 413.1 days. Inventory nearly tripled from USD 365.3 M in FY2023 to USD 1,046.0 M, and receivables went from USD 271.5 M to USD 610.7 M in the final year alone, because concentrate is being stockpiled for a smelter that is still ramping and refined output has only just started to sell. Operating cash flow was therefore negative in two of five years and free cash flow negative in three, roughly minus USD 1.6 to 1.8 B a year, funded by borrowings rather than by the business. Cash on hand still fell from USD 1,228.6 M at the FY2023 peak to USD 676.8 M.
Unit economicsThe cleanest way to see the transition is asset turnover. In FY2022 a dollar of assets produced 44 cents of revenue; in FY2025 it produced 13 cents, because total assets rose from USD 6.5 B to USD 13.9 B while revenue fell. That collapse is not decay, it is a balance sheet loaded with a smelter and a stripped pit that have not yet been asked to produce. Multiply it through and return on invested capital is 5.2% against a weighted cost of capital of 10.4%, so on the delivered record the company earns about half of what its capital costs. Everything therefore rests on one question that no ratio on this page can answer: whether the higher-grade centre of Phase 8, plus a smelter running near capacity instead of at 71%, lifts that 5.2% above 10.4%. The reserves and the mine plan are disclosed, so it is a checkable bet rather than a hopeful one, but it is still a bet on a future year.
AMMN owns the orebody via its IUPK (successor to the Newmont/Nusa Tenggara Contract of Work). Mining fleet, explosives, grinding media and reagents come from global OEMs and commodity suppliers on competitive terms. The 2025 smelter adds inputs (sulfuric-acid plant, oxygen, power) but AMMN built its own power plant to control the largest cost line.
Implication → Cost structure is grade- and strip-ratio-driven, not supplier-driven. Owning power generation vertically integrates the single biggest smelter operating cost: a deliberate margin-protection move.
Copper and gold sell at LME/LBMA benchmarks: AMMN is a price-taker. Pre-2025 concentrate sales gave smelters leverage via treatment/refining charges (TC/RC) that widened during the 2023–2024 global smelter overcapacity. In-country refining removes the TC/RC leakage but final metal is still benchmark-priced with no seller pricing power.
Implication → Integration recaptures the TC/RC margin (the strategic rationale for the smelter) but does not create pricing power: through-cycle returns still hinge on the copper and gold price decks.
World-class copper-gold porphyries are geologically rare and non-replicable: Batu Hijau and Elang are among the largest in Southeast Asia. Greenfield copper development runs USD 3–5 B+ and 10+ years of permitting; the IUPK and the sunk mine/mill/smelter/power complex are barriers no entrant can quickly clear.
Implication → The moat is geological + regulatory + capex. Domestic competition is limited to Freeport (Grasberg); AMMN is structurally the #2 Indonesian copper source with decades of reserve life.
Copper has no viable substitute in electrical, grid and EV applications: it is the core electrification metal, with structural demand growth into the 2030s (IEA). Gold has no industrial substitute as a store of value. Aluminium substitutes copper only at the margin in some cabling, insufficient to threaten demand.
Implication → AMMN sits on the right side of the energy transition: copper is a net electrification winner. This is the key strategic contrast with thermal-coal peers facing secular demand erosion.
Global copper is dominated by Codelco, BHP, Freeport, Glencore, Anglo: AMMN competes on cost-curve position, not scale. Domestically it sits alongside Freeport Indonesia (PTFI). Rivalry is expressed through cost efficiency and grade, not price (all output is benchmark-priced). Gold by-product credits are AMMN’s cost-curve advantage versus pure-copper miners.
Implication → Winning is lowest-cost extraction × favourable copper:gold grade × downstream margin capture. AMMN’s by-product gold keeps it low on the global copper cash-cost curve even in weak copper years.
The accounting is not the problem; the distance between profit and cash is, and it is the largest in this roster. Operating cash flow relative to net income ran 0.93×, 0.91×, −0.48×, 0.24× and −1.91×, so in FY2025 the company reported USD 249.0 M of profit while consuming USD 475.0 M of operating cash. The mechanism is visible in the same statements rather than hidden: inventory rose USD 274.6 M and receivables USD 339.2 M, together USD 613.8 M absorbed into working capital during smelter commissioning, when concentrate is stockpiled and refined metal has only begun to be sold. That is a legitimate explanation, and it is also a liability, because the profit is only validated when that stock converts to cash. Two further things belong in a reader’s head. First, depreciation has risen 150% to USD 323.3 M as the new assets are capitalised, so EBITDA-based measures increasingly flatter this company relative to operating profit, and the 54.0% EBITDA margin against a 36.5% operating margin is the gap to watch. Second, no one-off gains prop up any year here, and the effective tax rate is unremarkable, so the volatility in reported profit is genuine operating and grade volatility rather than accounting noise.
| Period | One-off item | Impact |
|---|---|---|
| FY2022 | The high-grade, high-price year: revenue +118% to USD 2,830.1 M on Phase 7 ore and a strong copper market. | Net income USD 1,093.5 M, ROE 31.0% and ROIC 26.0%, all the highest of the window by a wide margin. Grade and price are both cyclical inputs, so this is a peak rather than a baseline: measuring the later decline against it overstates how far the business has fallen. |
| FY2025 | The Phase 8 low-grade start plus smelter commissioning: concentrate output down 55% year on year in Q1, revenue −31%, and USD 613.8 M absorbed into inventory and receivables. | ROE 4.7%, ROIC 5.2% and operating cash flow of minus USD 475.0 M. This is a transition year in both the mine and the plant simultaneously, so it is the wrong year to annualise in either direction, up or down. |
Cash conversionWeak by design right now, and the design is the point. Operating cash flow ran USD 293.6 M, 990.5 M, −121.2 M, 151.7 M and −475.0 M against capex of USD 154.8 M, 692.3 M, 1,519.6 M, 1,791.8 M and 1,372.4 M, so free cash flow was roughly minus USD 1.6 B, 1.6 B and 1.8 B in the last three years. Cumulatively across FY2023 to FY2025 the business generated minus USD 444.5 M of operating cash while spending USD 4,683.8 M on capital projects, which is why borrowings had to rise USD 4.6 B. None of that is disguised, and the assets bought are disclosed, but it does mean this company has not funded itself from operations for three years.
The most aggressive builder in the roster measured against its own size, and the one where the outcome is genuinely undecided. Across FY2023 to FY2025 it spent USD 4,683.8 M of capex while generating minus USD 444.5 M of operating cash, so essentially the entire programme was financed externally: borrowings rose from USD 1,862.6 M to USD 6,461.6 M and no dividend has been paid. Judged the only way that settles the question, the record says the capital is not yet earning its keep: ROIC of 5.2% against a 10.4% cost of capital, and below that cost in three of the five years. The counter-case is specific rather than vague, which is what makes this analysable at all: the spending bought a 900,000 tpa smelter now at 71% utilisation, a Phase 8 pit whose higher-grade centre is still ahead, roughly 460 Mt of added reserves and a mine life to 2030, plus a completed Elang feasibility study. Those are checkable milestones. Contrast DCII, which builds at 19.1% ROIC against a 12.2% cost of capital and therefore compounds while it spends; AMMN has to reach that condition before it can compound.
DeploymentWhere the money went, in order of size. Into capital projects: USD 154.8 M, 692.3 M, 1,519.6 M, 1,791.8 M and 1,372.4 M, dominated by the smelter and precious-metals refinery and by Phase 8 stripping. Into working capital: USD 613.8 M in FY2025 alone as inventory and receivables built. To lenders: interest of USD 386.0 M in FY2025, up from USD 119.3 M in FY2021, on borrowings that more than tripled. To shareholders: nothing, in any year. Funded by: debt, plus a drawdown of cash from USD 1,228.6 M to USD 676.8 M. Total assets rose from USD 5.2 B to USD 13.9 B over the window, so this is a company that has roughly tripled its balance sheet in five years while its revenue ended lower than it started.
Returns trendROIC 10.6%, 26.0%, 5.1%, 9.5%, 5.2% against a weighted cost of capital of 10.4%, so the delivered record clears the bar in only one year, the FY2022 peak. ROE tells the same story: 13.1%, 31.0%, 5.5%, 12.3%, 4.7%. On valuation, this page deliberately shows no per-share DCF figure, and the reason is worth understanding rather than skipping. On the default normalised assumptions the modelled enterprise value falls below net debt of about USD 5.8 B plus minority interests, so the equity residual is negative, and equity cannot be worth less than nothing. Publishing a negative price would be arithmetic theatre, so the model states the condition instead: at this discount rate, on cyclically normalised margins, the debt claims consume the whole modelled enterprise. That is a statement about a mid-transition record, not a prediction of failure, and it is precisely why the multiples matter more here than the DCF: 66.0× earnings, 22.4× EV/EBITDA and a free-cash-flow yield of MINUS 11.2% say the price is already paying for the finished asset. The honest way to hold this name is as a milestone bet: smelter utilisation above 71%, the Phase 8 grade ramp, and the point at which operating cash flow turns positive again.
Operating cash flow was minus USD 475.0 M against reported profit of USD 249.0 M in FY2025 (OCF/NI −1.91×), and negative in two of the five years. The cause is identifiable, USD 613.8 M absorbed by inventory and receivables during commissioning, and the cash-conversion cycle stretched to 413 days. It is explainable, but the profit is only proven once that stock sells. Watch inventory and the cycle before watching earnings.
Borrowings more than tripled to USD 6,461.6 M, debt-to-equity reached 1.19, and interest cover fell from 10.76× in FY2022 to 1.74×, with interest now consuming 20.9% of revenue. Fixed capital commitments sitting on revenue that swings with ore grade is the combination that removes a company’s room for a disappointing year. This is the mirror image of BYAN, which entered its downturn with no debt at all.
ROIC of 5.2% against a 10.4% WACC, below cost of capital in three of five years, while invested capital itself is rising fast (total assets USD 5.2 B → 13.9 B). Adding capital at a sub-cost return compounds the shortfall rather than curing it, so the entire investment case depends on the Phase 8 grade ramp and smelter utilisation lifting that number, not on the copper price alone.
Batu Hijau is effectively the whole company until Elang is developed, its Phase 8 mine life runs to 2030, and the smelter that changes the revenue mix was still at roughly 71% utilisation (55% at the precious-metals refinery) at the end of 2025. Concentrate export permits were extended to bridge that ramp, which is itself a dependency on policy. One pit, one plant, one permit regime.
No concern in the books themselves: no one-off gains support any year, the effective tax rate is unremarkable, reserves and grades are disclosed, and the working-capital swing that drove cash negative is visible line by line. The caution is interpretive rather than forensic, namely that rising depreciation (USD 129.4 M → 323.3 M) makes EBITDA-based measures increasingly flattering relative to the 36.5% operating margin.
Curated narrative (educational interpretation), backed by the linked sources. The figures above are deterministic. Not investment advice.
Indonesia controls >60 % of global nickel supply and is the world's largest thermal-coal exporter; hilirisasi mandates in-country ore processing, reshaping value chains from raw ore to battery-grade products.