Ivanhoe Electric: the stock already prices copper above the banks' long-term forecasts

On September 23, 2026 Ivanhoe Electric published a new study for its Santa Cruz copper project, and the headline number was big: at $6.79 copper the mine is worth $3.5 billion after tax, more than twice the company's $1.54 billion market value. The same study raised the initial construction cost by 15% to $1.426 billion and pushed first copper a year later, to 2029. At the company's own $4.75 copper price the mine is worth $1.519 billion, which is almost exactly what the stock market pays for the whole company today. The cheap-against-NPV story only works if you value the mine at spot copper.

At the October 8, 2026 close of $9.66 the shares trade 31% above my triangulated Base value of $7.38, a downside of 23.6%. The Bear value is $0.93, a 90.4% drop, and the Bull value is $11.35, 17.5% above the price. The required margin of safety (MOS) is 80%, above the framework's 60% ceiling, so I don't issue a buy price at all. Even after a final federal loan commitment the required margin would be 70%, still above the ceiling.

This call sits close to a line, and I want to say so up front. The Base is 23.6% below the price, which my rules classify as Slightly Overvalued, a band where both HOLD and REDUCE are allowed. HOLD is a reasonable alternative reading of the same numbers. It becomes the call if the Export-Import Bank of the United States (EXIM) gives a final commitment of at least $1.0 billion with capex at or below $1.43 billion, if long-term copper expectations reach about $5.41 a pound, or if a discount rate of 9.2% or lower becomes defensible. I chose REDUCE because the Bear is -90.4% against a Bull of +17.5%, Confidence is D, and the last study moved both cost and schedule the wrong way.

Route: SF-12 Mining. Subtype: pre-revenue, single-asset copper developer (underground mine producing copper cathode in Arizona) with an exploration and technology portfolio. Economic classification: Standard (2.4/12). Primary method: risked net asset value (NAV), the present value of the mine's after-tax cash flow after construction and sustaining capital, less corporate costs, plus net cash. Secondary method: enterprise value (EV) as a multiple of EBITDA (earnings before interest, taxes, depreciation and amortization) at a mid-cycle copper price, discounted to today. Values are present fair value estimates, not 12 month price targets.

What Ivanhoe Electric is

Ivanhoe Electric is a copper developer with one asset that matters: Santa Cruz, near Casa Grande, Arizona, 100% owned, on private land. The 2026 preliminary feasibility study (PFS) shows 140.1 million tonnes of probable reserves at 1.08% copper, a 24-year mine life and 74,722 tonnes a year of copper cathode for the first 15 years. The mine is underground, and the copper is leached and recovered into cathode on site by solvent extraction and electrowinning (SX-EW), so it sells a finished metal inside the US rather than concentrate. A royalty of 1.68% goes to Deterra until the sixth anniversary of commercial production, then 1.57%.

Revenue is incidental, under $1 million a quarter from geophysics and battery services. Around the mine sits a portfolio: a 50/50 exploration joint venture with Ma'aden in Saudi Arabia, the Tintic, Hog Heaven and Gleeson projects, 60.8% of Cordoba Minerals (mostly cash since it sold its Alacran stake for $128 million), VRB Energy (vanadium flow batteries) and Computational Geosciences, which runs the Typhoon geophysics system. The executive chairman, Robert Friedland, owns about 7.3%, and Ma'aden about 7.4%.

The plan from here: a tunnel boring machine starts the decline in summer 2027, EXIM's board is expected to decide on up to $1.1 billion of project debt in spring 2027, and first cathode is scheduled for 2029. City, county and state permits for surface construction and the decline are reportedly in hand.

The latest quarter

The second quarter was a developer's quarter: almost no revenue, a modest loss, and a balance sheet held up by asset sales and warrant exercises rather than by operations.

The first half looks profitable on paper and isn't. The Alacran gain is a one-off, part of it belongs to Cordoba's minority shareholders, and $40.1 million of the cash already went to them. Strip it out and the run rate is a corporate and exploration burn of roughly $25 million a quarter, before construction spending starts.

What the September study changed

The 2026 study replaced the 2025 one, and comparing the two side by side is the most useful thing in the filings, because the two studies are the only public commitments management has made on this project.

The headline is the higher NPV (net present value). The table shows where it came from: the copper price assumption rose 50 cents a pound, while capex, cash cost and the schedule all moved against the project. Per share, with the share count up about 15% a year, the study value did not grow. The NPV went up because copper did.

That NPV is also not a value per share. It is an asset-level number at an 8% discount rate and a flat $4.75 deck, before corporate overhead, stock-based compensation (SBC), exploration spending and the cost of financing a $1.426 billion build. My NAV on a higher copper deck and a 10% discount rate is $1,280.7 million, 84.3% of the study's headline. Market value is 1.01x the study NPV and 1.20x my NAV at 10% (1.23x per diluted share).

Cash, funding and the share count

Net cash, defined once and used everywhere in this article, is $197.1 million: $256.9 million of cash, less $23.5 million held in subsidiaries that are not wholly owned, less the $35.2 million VRB Energy convertible bond due December 31, 2026, less $1.1 million of leases. The $200 million bridge facility at the project company was undrawn at June 30. It is not free money: Ivanhoe Electric guarantees it, the lenders hold a lien on all of the project company's assets, the rate steps up 0.5 percentage points every six months, and the parent must keep a tangible net worth of at least $225.0 million. The current headroom against that covenant is not in the data available to me.

Pre-production cash needs from July 1, 2026 are about $1,671 million: the $1,426 million build plus about 3.5 years of corporate costs and exploration. Against that sit $197.1 million of net cash and up to $1.1 billion from EXIM, which is a preliminary letter, not a commitment. That leaves an equity gap of $373.9 million even if EXIM comes through in full. Without EXIM the gap is $1,473.9 million, 0.96x today's market value.

The share count went from about 103.7 million in 2023 to 159.59 million in August 2026, a 53.9% increase, or 14.9% a year. All 11.6 million warrants at $7.00 were exercised in January and February 2026 for $81.5 million, and Ma'aden has a right to top up to 9.9% on every future issue. I use 162.59 million diluted shares, which adds an estimated 3.0 million options and stock units to the reported count.

The valuation divides NAV by today's diluted shares, which is the same as assuming the equity gap is raised at fair value. Raising it at today's price, above my Base, would lift the Base to $7.62, but a value that depends on the market overpaying for new shares is not intrinsic value, so it is a robustness check only.

Primary valuation: risked NAV

I rebuilt the project's cash flows from the published study parameters: production, recovery, cash cost, sustaining capex of $1.55 billion over the mine life, initial capex spread from late 2026 to 2029, an 18% effective cash tax and real (inflation-adjusted) prices. The full technical report model is not used, so I calibrated the rebuild to four published outputs.

The rebuild sits slightly below the study. Scaling it up would add $0.34 to the Base, and I don't do it.

The copper deck is the input that decides everything. The study uses $4.75 flat, which is below spot, below the forward curve and below the bank long-term forecasts, so it is not the right Base. My Base deck is $6.00 a pound in 2029 and 2030, falling in a straight line to $5.20 by 2033 and flat after that. The $6.00 is a real-dollar figure: forwards for 2029 and 2030 trade at roughly $6.4 to $6.7 a pound, which is about $5.9 to $6.2 in today's dollars, and UBS's medium-term forecast is $6.00. The $5.20 long-term level is the low end of the bank long-term decks I found (Goldman Sachs $5.20, BofA $5.44, UBS $5.50). The deck is my estimate, not a downloaded futures curve.

That places the deck in the middle of the evidence. It is above the company's $4.75 study deck and the $4.94 market long-term consensus from the end of 2025, at the low end of the bank long-term decks, and roughly at the forward curve in today's dollars. Spot copper peaked at about $6.85 to $6.95 a pound in September and was about $6.49 on October 2, so the Base long-term level is about 25% below the latest spot print.

The discount rate is 10% real, the middle of the 8% to 12% range the framework sets for development projects. The risk discount goes in once, through the rate, and not again through a haircut on NAV.

SBC is deducted because it is a real cost paid in shares. Leaving it out would give a Base of $7.95, still below the HOLD line.

Sensitivity: primary value per share

Near-term deck at $6.00 in 2030, fading in a straight line to the long-term price by 2033. The $6.00 column is flat.

The grid shows why a mining NAV needs a wide margin of safety. Two percentage points on the rate move the value by $2 to $3 a share, and the difference between $4.75 and $6.00 long-term copper is $6.01 against $11.19.

Secondary valuation: mid-cycle EBITDA, discounted

The second method values the running mine at the start of its first full year, 2031, as a multiple of EBITDA at the mid-cycle copper price, adds the after-tax premium of the early high-price years, subtracts every cash need between now and then, and discounts everything to today at the same 10%.

This method is not fully independent. It shares production, costs, the capex schedule, the discount rate and the deck with the primary. What differs is the mechanism: one year's earnings times a market multiple instead of 24 years of explicit cash flow. No peer price-to-NAV or EV-per-pound table is used, so the 5.5x is the middle of the framework's sector range rather than a peer median.

The secondary comes out lower because a single-year multiple undervalues a long mine life. To match the primary the multiple would have to be about 6.3x, still inside the sector's fair range.

Sensitivity: secondary value per share

History and growth diagnostics

There is no usable history of price to NAV, because the share count, the study and the copper price all changed inside the last year. As a proxy, market value on today's share count against the $1,519 million study NPV ranged from 0.83x to 2.26x across the 52-week range of $7.86 to $21.55. Today it is 1.01x, in the lower part of that range. Against my NAV on the Base deck at 8% the price is 0.91x per diluted share (0.90x on basic market value), inside the sector's fair band of 0.8x to 1.2x. At 10% it is 1.20x basic and 1.23x diluted, the top of that band. This layer gets no weight.

The growth cross-check (SM-19) does not produce a usable PEG ratio: there are no earnings, the second quarter loss was $0.16 a share, and the first quarter was distorted by the Alacran gain. Free cash flow (FCF) was negative in all three available years: -$232.6 million in 2023, -$175.7 million in 2024 and -$94.3 million in 2025. The improvement came from lower exploration and capex, not from operations, so a growth rate on it means nothing.

The FCF quality check (SM-20) is therefore a path-to-cash-flow exercise. In the first full year, 2031, on the Base deck, the mine would produce about $422.6 million of FCF after sustaining capex, interest on $1.1 billion of EXIM debt at an estimated 7% and tax. After raising the $373.9 million equity gap at today's price that is $2.10 a share, 21.7% of today's price, five years out. At the $5.20 mid-cycle price it is $1.74 a share, 18.0%. Discounted at 10%, the 21.7% is about 13.8% in today's money, before construction and price risk. Reliability: low. Neither diagnostic sets the Base, the margin of safety or the verdict.

Valuation triangulation

The NAV gets the larger weight because it uses the whole mine life and is calibrated to the published study. The two methods are 23.4% apart, inside the 20% to 30% warning band, and the Base is not set at the higher of the two. The gap is fragile: with the rebuild scaled to the study it is 29.2%, and at an 8% discount rate it is 39.8%, because the NAV is much more rate-sensitive than a single-year multiple. Even if that 8% figure triggered the divergence penalty, Confidence would be 6 (D) and the required MOS 85%, and the outcome would be the same: no buy price.

What is the market already pricing in?

Running both methods backwards from $9.66, with the same net cash and the same share count, gives the copper price or discount rate the market is paying for.

At $9.66 the market pays for the forward curve to 2030 and then for long-term copper of about $5.79 a pound, above the $5.20 to $5.50 bank long-term decks and 17% above the $4.94 market long-term consensus. Or it pays for a 7.9% discount rate, the bottom of the development range, on a mine that is neither financed nor built. Spot at $6.49 shows the copper belief isn't unreasonable today. My Base simply doesn't extend spot beyond the forward horizon.

Bear, base and bull scenarios

One lever moves between the scenarios: the copper deck. Capex, schedule, the 10% rate, the 5.5x multiple, corporate costs and the share count stay the same, so the table shows copper risk on its own. The multiple is deliberately not cut in the Bear, because a trough copper price usually comes with a higher multiple, not a lower one, and cutting both would count the same bad news twice.

The Bear is about 38% below spot and 23% below the Base long-term level. The Bull is above spot in the early years and above every bank long-term deck I cited after 2033, and it still gives only 17.5% of upside. I assign no probabilities to the three cases.

Each 25 cents on long-term copper moves the Base by about $0.97. The near-term level matters much less: each 50 cents on the 2029 to 2030 price moves it by about $0.43. If you plug in the nominal 2029 to 2030 forwards directly, $6.37, $6.49 or $6.73, the Base becomes $7.70, $7.81 or $8.01 (-20.3%, -19.2% or -17.0%). The classification doesn't change. HOLD needs a near-term level of $6.96. Held flat for the whole mine life, spot copper of $6.49 to $6.79 would give $12.82 to $14.24.

A floor check: if the mine were never built, net cash less two years of overhead and holding costs is worth about $0.72 a share. The Bear is above that.

How the buy price is set

A Confidence D score starts the required MOS at 50%. Four risk modifiers come on top.

80% is above the 60% practical ceiling, and at that point a buy price would be a precise-looking number with no real meaning. So the framework issues none: no buy zone, no maximum buy price. The stock would not qualify as a buy even at the Base value. A final EXIM commitment would remove the financing modifier, and 70% is still above the ceiling.

Dividend and capital return

No dividend and no buyback. The cash yield is 0%, and dividend coverage tests do not apply. The only distribution in the period was at subsidiary level: Cordoba paid $40.1 million to its minority shareholders in the first half of 2026, which is cash that left the group, not cash returned to Ivanhoe Electric's shareholders.

Reasons to own IE

• A rare US copper asset. 1.08% copper reserves, 100% owned on private land in Arizona, a 24-year life and cathode produced inside the US, where copper on COMEX, the US futures exchange, has traded at a premium. Few comparable US projects exist.
• Copper leverage. Each 25 cents a pound on long-term copper adds about $0.97 a share, and spot copper held flat gives $12.82 to $14.24.
• A team that raises and recycles capital. The chairman's group has built large copper mines before. The Alacran sale brought in $128 million, and the $200 million bridge is in place and undrawn.
• A financing path. EXIM's preliminary letter for up to $1.1 billion, with a board decision expected in spring 2027, would cover about 77% of initial capex with long-tenor debt.

What could go wrong?

• Thesis breaker: copper. The Base already assumes the forward curve to 2030 and $5.20 long term. At the company's own $4.75 deck the Base is $4.53, at the $4.94 market long-term consensus with the same near-term path it is $6.37, and at $4.00 flat it is $0.93.
• Funding. About $1.67 billion of pre-production needs against $197.1 million of net cash. Without EXIM the gap is $1.47 billion, 0.96x today's market value. The parent guarantees the bridge and must keep a tangible net worth of at least $225.0 million.
• Construction and cost. Capex already rose 15% and first cathode slipped a year between studies. Another 15% on capex costs $1.04 a share, and another year of delay $1.33. Either one alone takes the Base below the -30% line.
• Dilution. The share count grew about 15% a year. The equity gap and Ma'aden's top-up right make further issuance certain, not possible.
• The VRB bond. $35.2 million is due on December 31, 2026, after its maturity was already extended once.

Management execution

Management has delivered on financing. It closed a $172.5 million public offering in October 2025 and the $200 million bridge in December 2025, collected $81.5 million from warrants in early 2026, raised the EXIM letter from $825 million to $1.1 billion, sold Alacran, VRB China and Pinaya, and made the final Santa Cruz option and land payments. Against that, the two studies are the only public commitments on the mine, and both moved the wrong way between 2025 and 2026: capex from $1.236 billion to $1.426 billion, C1 cost from $1.32 to $1.47 a pound, first cathode from 2028 to 2029. The side ventures, VRB and CGI, consumed cash for years. Assessment: 9/20, mixed. The next test is the third quarter 10-Q, expected around November 5, 2026, with the covenant headroom, the bridge balance and the cash position.

Stock Analyza scorecard

Business quality is 34/100: asset quality and cost position 9/20 (good grade, private land, US cathode, but a price taker with an unbuilt mine and rising costs), return versus cost of capital 4/20 (no realized return yet), balance sheet 4/15 (net cash today, but the build is unfunded and the bridge carries a covenant), earnings and cash flow quality 2/15, growth and runway 8/15, management and capital allocation 5/10, dilution and alignment 2/5. The valuation score is 28/100: discount to Base 11/40 (the Base is 23.6% below the price), protection against the Bear 0/20, agreement between methods 7/15, historical valuation 5/10, reverse valuation 4/10 (the implied long-term copper price is 11% above my Base deck, but below spot), data and model quality 1/5.

Confidence score

Data quality is low because the copper deck is my estimate built from bank and market commentary, the project model is a rebuild, filing figures were taken from published summaries and excerpts rather than reproduced line by line, and the option overhang and the futures curve beyond 2028 are not available. Valuation robustness is 1 because several single inputs cross a classification line within their normal range. A range from $0.93 to $11.35 is what holds scenario dispersion at 1.

What I would watch from here

For a REDUCE call the colors describe the thesis, not the share price. Green means the REDUCE case holds. Red means it breaks and the call moves toward HOLD, so a rising copper price shows up as Red here, and that is good news for the stock. A separate escalation list below covers the opposite direction, toward SELL.

Escalation toward SELL, the opposite direction: long-term decks below $5.04 or 2029 to 2030 forwards below $5.28 (the Base falls below the $6.76 line), capex above $1.55 billion, first cathode a full year or more later, no EXIM commitment by the end of 2027 or less than $800 million, net cash below $100 million without new financing, covenant headroom below $50 million or a waiver, the bridge fully drawn with EXIM delayed, a default or recourse to the parent on the VRB bond, more than 185 million shares or any raise below $8.00, and quarterly burn above $40 million.

Today no reading is Red. Near-term copper is Yellow: spot was about $6.49 on October 2, between my $6.00 level and the $6.96 Red line, after a September peak of about $6.85 to $6.95. The VRB bond is Yellow until it is settled on December 31. Covenant headroom is treated as Yellow until the third quarter 10-Q discloses it. Everything else is Green: the long-term deck, capex, schedule, EXIM on track, net cash ($197.1 million at June 30 and an estimated $162 million today), the undrawn bridge, the share count and the burn. The next review point is the third quarter 10-Q, expected around November 5, 2026.

The three most important thesis breakers

1. Long-term copper expectations at or above $5.41 a pound with 2029 to 2030 forwards near $6.00, or 2029 to 2030 forwards at or above $6.96 for 12 months or more with the long-term level unchanged.
2. A final EXIM commitment of at least $1.0 billion with capex at or below $1.43 billion. With that de-risking a 9% discount rate becomes defensible, and 9% alone gives $8.39, inside the HOLD band.
3. Capex at or below $1.26 billion, a reserve or mine life extension, or other assets worth more than about $235 million. Each of these is most likely to matter in combination with 1 or 2.

Adversarial review

I recomputed every value in this article in a separate model and checked the inputs against the second quarter 10-Q, the 2025 10-K, the September 2026 study release and its 8-K, published summaries of the study and the closing price from two independent market data sources. The balance sheet, share count, bridge status, covenant, ownership and study parameters are reported figures. The copper deck, the discount rate, the 5.5x multiple, the $100 million for other assets, the 3.0 million share overhang, the exploration budget and the 7% interest rate on project debt are my estimates.

The weakest point of the model is the copper deck. No futures curve beyond 2028 was used, the bank forecasts mix real and nominal figures, and the model works in real dollars. The second weakest is the rebuilt project model, calibrated to within 5.0% of the study but not the study's own cash flow file. The third is that no single non-copper input stays neutral across its plausible range: a 9% rate, a 7x multiple or $250 million of other assets each reach HOLD, while an 11% rate, 15% more capex or a one-year delay each reach the -30% line.

The strongest counter-thesis, for the bulls: the market prices consensus, not euphoria. Banks see about $6.00 copper to 2029 and $5.20 to $5.50 long term, the asset is 100% owned on private land in Arizona with local permits in hand, EXIM has signaled $1.1 billion and the bridge is undrawn. Combine the nominal forward ($6.49) with the bank long-term average ($5.38) and the Base is $8.50, only 12.0% below the price, which is a HOLD. The response is that this combination mixes a nominal forward with a real model. In today's dollars the near-term level is about $5.9 to $6.2 and the Base $7.26 to $7.55. Even the bull reading ends at HOLD, not BUY: the BUY line of $11.11 needs long-term copper of $6.13. Outside the deck, capex rose 15% and first cathode slipped a year in the last study cycle, the equity gap is $374 million even with EXIM, and the share count grows about 15% a year.

The strongest counter-thesis, for the bears: the company's own deck is $4.75, and on it the Base is $4.53, -53.1%. The market long-term consensus of $4.94 gives $6.37. The response is that the framework sets the Base at the forward and consensus deck, not the company's study deck, which sits below spot, the forwards and every bank long-term forecast. Both bear decks are shown, and both say Highly Overvalued.

The verdict is a centered estimate inside a wide band, which is exactly why Confidence is D and the required MOS is 80%. Arithmetic: 0.7 times $7.877 plus 0.3 times $6.229 is $7.3825; $7.3825 divided by $9.66, less 1, is -23.6%; the Bear is 0.7 times $1.084 plus 0.3 times $0.559, or $0.93; the Bull is 0.7 times $12.055 plus 0.3 times $9.721, or $11.35; the divergence is $1.648 divided by $7.053, or 23.4%; the MOS is 50 plus 10 plus 10 plus 5 plus 5, or 80%; market value is $9.66 times 159.59 million shares, or $1,541.6 million. The publication gate status is PASS WITH WARNING: the arithmetic reproduces, and the value rests on an estimated copper deck, a rebuilt project model and an undisclosed covenant headroom.

Final verdict: reduce at $9.66

Santa Cruz is a real asset: good grade, 100% owned, on private land in Arizona, producing cathode inside the US. None of that is in dispute. What the price asks you to accept is long-term copper of about $5.79 a pound, above the banks' long-term decks, on a mine whose construction cost rose 15% in a year, whose first cathode moved to 2029, and whose $1.1 billion of debt is still a preliminary letter. HOLD is a defensible alternative, and the triggers that would make it the call are listed above: a firm EXIM commitment at today's capex, long-term copper expectations at $5.41 or higher, or a case for a discount rate of 9.2% or lower.

Verdict: REDUCE at $9.66, do not add. Triangulated Base value $7.38, downside 23.6%, Confidence D (7/25). Required margin of safety 80%, no buy price issued.

The study's $3.5 billion is real at $6.79 copper. The share price skips the question of how much of it is left for today's shareholders once a $1.426 billion mine has been paid for.

Sources

Primary sources include Ivanhoe Electric's Form 10-Q for the quarter ended June 30, 2026 (cash, subsidiary cash, the VRB Energy bond, the undrawn bridge facility, the guarantee and tangible net worth covenant, cash flows and the share count), the Form 10-Q for the quarter ended March 31, 2026 (stock-based compensation), the 2025 Form 10-K and the 2024 Form 10-K, the 2026 proxy statement, the company's 2026 preliminary feasibility study release and its 8-K summary, the announcement of 100% ownership of the Santa Cruz mineral rights, the bridge facility closing, the October 2025 public offering and the Alacran amendment 8-K. Study details, sustaining capex, schedule and permits come from Mining.com, The Northern Miner, Mugglehead, The Deep Dive and Redimin, and the royalty terms from Deterra Royalties' ASX announcement. Copper deck evidence comes from Mining.com on long-term price assumptions, S&P Global Market Intelligence on feasibility study price decks, Hellenic Shipping News on BofA's long-term forecast, Tiger Brokers, ExchangeRates.org.uk on medium-term forecasts and deVere Group. Spot copper comes from Barchart, AlphaMaven and KuCoin News. The closing price comes from Yahoo Finance and MarketChameleon.

The reference market price is the October 8, 2026 regular session close of $9.66 on NYSE American. Balance sheet figures are as of June 30, 2026, and the share count of 159,590,259 is as of August 7, 2026. Third quarter results and the current covenant headroom were not available. The copper deck, the discount rate, the EV/EBITDA multiple, the value of other assets, the share overhang, the exploration budget and the interest rate on project debt are analytical estimates, not company guidance. No peer price-to-NAV or EV-per-pound table is used; the multiple range is the framework's sector range.