Megawatts leased to AI tenants currently trade at a different price on Wall Street than megawatts in the pipeline of Bitcoin miners, and the distance between the two has become a central issue in pricing across the sector.
VanEck’s latest framework for valuing publicly traded miners shows that companies with AI and high-performance computing lease agreements trade at more than 10 times their total energy production, while miners with little or no contracted capacity trade at roughly 2-6 times that metric.
Investors are starting to treat leased megawatts as a more distinct and valuable asset class than mined Bitcoin or unsold power capacity.
| metric | van eck diagram | why is it important |
|---|---|---|
| Miners with AI/HPC leases | More than 10 times the total energized power | Wall Street is assigning a premium to contracted AI capabilities |
| Miners with little or no contracted capacity | Approximately 2x to 6x total energized power | The value of the pipeline alone is much lower than the signed lease |
| AI/HPC capacity provided | Up to 25% of leased capacity | Most of the contracted capacity still needs to be built and delivered |
| Short-term funding shortfall | ~$50 billion | This sector requires significant capital before leases become cash flows |
| Long-term capital will be required if the pipeline is converted | ~$221 billion | AI Pivot Could Become an Infrastructure-Scale Funding CycleA |
Premium will arrive before capacity is reached
VanEck puts the AI and HPC capacity provided across its peer group at only about 25% of its leased capacity. Today, Wall Street is paying for contracts and for construction outcomes that the sector has yet to realize.
The short-term funding gap for its construction amounts to around $50 billion across the group, with long-term financing needs rising to $221 billion if the entire pipeline of announced projects is eventually converted into construction land.
VanEck’s valuation model assumes a baseline net operating income of approximately $1.5 million per megawatt for AI and colocation sites and applies an enterprise value of 15 times that number.
This model also offsets results against greenfield construction costs of about $10 million per megawatt, rising to about $12 million for projects further down the road with additional construction inflation.
One megawatt represents a total enterprise value of nearly $22.5 million, compared to a pre-capex value of approximately $12.5 million, before stochastic discounts for delivery risks and financing costs are applied.
| input | prediction | implicit value |
|---|---|---|
| Net operating income per MW | ~$1.5 million | Starting cash flow base |
| Corporate value multiple | 15 times | Convert NOI to asset value |
| Total enterprise value per MW | $1.5 million x 15 | ~$22.5 million |
| Greenfield construction cost | ~$10 million/MW | Baseline capital investment deduction |
| Value after capital investment and before financing | $22.5 million – $10 million | ~$12.5 million |
| Capital investment for further projects | ~$12 million/MW | As costs rise, the implied equity value falls. |
| Main sensitivity | Capital investment, timing, tenant quality | Small changes can make a big difference in shareholder value |
If the capital investment per megawatt increases by a few million dollars or the delivery date is extended by a year, the capital value attached to that megawatt changes proportionately and significantly.
VanEck’s framework assumes that megawatts leased to investment-grade hyperscalers can be supported with an effective cost of capital of 6% to 10%. Leasing similar megawatts to smaller GPU cloud tenants can guarantee discount rates in excess of 10%, and the cost of capital increases directly with the tenant’s risk.
Contracted leases and energized megawatts have different values when considering a tenant’s balance sheet. If you sell the same power to a weaker trading partner, the premium will be smaller.
Make up for the shortfall without giving room for growth
By closing the $50 billion short-term shortfall, miners will be drawn to funding tools from infrastructure and project finance.
Project finance and debt bring fixed debt to balance sheets built around volatile mining margins. Selling Bitcoin bonds would turn assets that some miners have spent years accumulating into construction funds, undermining the initial theory that drew investors who focused on Bitcoin into the stock in the first place.
Strategic partnerships and tenant prepayments offer a softer path, but typically come with conditions that shift some of the AI-era upside from existing shareholders to the partner providing the capital.
The International Energy Agency predicts that global data center power consumption will nearly double from about 485 terawatt-hours in 2025 to about 950 terawatt-hours by 2030, and AI-specific data center consumption will triple over the same period.
McKinsey predicts that global data center spending could reach approximately $7 trillion by 2030, with approximately $5.2 trillion of that going to AI-enabled facilities.
KKR recently launched a $10 billion AI infrastructure venture with Nvidia, and Vistra shows that major financial institutions are treating power-backed AI capabilities as their own asset class, expanding capital at a pace that matches the size of the opportunities miners are pursuing.
Bitcoin’s shadow has not disappeared
Even though VanEck’s framework describes a business model that transitions to AI leasing, the market continues to price miners based on Bitcoin’s daily fluctuations.
The group’s one-year average weekly beta for Bitcoin is close to 1.05, meaning the typical mining stock is still moving roughly in lockstep with Bitcoin’s price, even as its underlying cash flow story has shifted to AI leasing.
Meaningful Bitcoin government bond exposure that would justify beta is concentrated in a small number of names.
| Company/Organization | BTC holdings (percentage of market capitalization) | what it suggests |
|---|---|---|
| Mara | ~51% | Still has a meaningful connection to Bitcoin’s financial value |
| CLSK | ~24% | BTC exposure remains important |
| riot | ~11% | Some links with BTC balance sheet |
| hut | ~7% | Limited but visible BTC exposure |
| most other peers | ~1% or less | BTC Beta May Overestimate Actual Balance Sheet Exposure |
| Peer group average beta to BTC | ～1.05 | Stock prices still trade at roughly 1:1 with Bitcoin |
MARA holds Bitcoin worth about 51% of its market capitalization, CLSK about 24%, RIOT about 11%, and HUT about 7%, while most of its peers hold Bitcoin worth less than 1% of their market capitalization.
Winners focused on AI may trade too cheaply during Bitcoin declines, while laggards focused on pipelines may trade too richly whenever Bitcoin rises.
VanEck’s Governance Scorecard measures insider ownership, management KPIs, executive compensation structure, management tenure, and related party transactions, and no company in the group scores close to a perfect score, with HIVE and BTDR ranking low on the relative scale.
Funding tens of billions of dollars for AI infrastructure will require investors to trust management teams with capital budgets orders of magnitude larger than what mining-era balance sheets previously required.
Governance gaps had little impact in the hashrate business, but had a real impact in the business that sells power to hyperscalers under long-term contracts.
Two paths from contract to cash flow
The bullish case for the sector is that miner valuations move into a framework already used by data center REITs and infrastructure landlords.
Hyperscaler demand for power-dense, interconnect-ready sites remains strong, financing markets are opening up to credit-worthy projects, and miners furthest along in construction are beginning to report megawatts delivered and recurring lease revenues.
Capacity at the time of delivery remains close to or above 10 times what VanEck has already observed, and the premium assigned by the market early will ultimately be validated by subsequent cash flows.
In the bearish case, the funding gap is resolved through dilution as rising labor, equipment, and grid interconnection costs raise construction costs beyond the $10 million per megawatt threshold.
Debt is a price for a sector with limited history as an infrastructure landlord, prompting miners to turn to equity issuance and Bitcoin monetization to make up for the shortfall before AI returns materialize.
Shareholders fund the build-up, with a significant portion of the final appreciation going instead to lenders, strategic partners, or buyers of newly issued shares that set the entry price after dilution.
The tests that determine which cases unfold have nothing to do with the scale of miners’ next AI announcements.
Ultimately, this will determine the number of megawatts delivered relative to the megawatts leased, the creditworthiness of the tenant signing each lease, and the actual capital investment required per megawatt once construction begins.
It also depends on the financing structure chosen to bridge the distance between today’s cash and tomorrow’s revenue, and whether each company’s governance can support capital allocation at infrastructure scale.
Wall Street has already decided that these companies are more valuable as AI infrastructure than Bitcoin miners.
What remains unresolved is whether investors are paying for AI cashflow that has yet to materialize, or whether they are paying for a construction pipeline that still requires tens of billions of dollars without any AI revenue.
(Tag translation) Bitcoin
