Building DivGalaxy Digital: When Two Supply Constraints Intersectidend Portfolios

Galaxy Digital operates at the intersection of two markets experiencing fundamentally different supply dynamics. The company generates fees from crypto trading activity that responds immediately to volatility, while simultaneously controlling secured power capacity for data centers at a site where replication takes years and regulatory approval remains uncertain. The mispricing exists in treating these exposures as redundant rather than complementary. Crypto trading produces cash flow that scales with activity cycles. Data center power produces contracted revenue from capacity that cannot be replaced once demand arrives. The setup does not require synchronized timing. It requires recognizing that cash-generative businesses and infrastructure bottlenecks compound differently under constraint.

The year since August 2024 has compressed what would normally unfold across market cycles into overlapping waves of scarcity. Crypto markets entered a period of renewed institutional participation through ETF vehicles while simultaneously confronting regulatory clarity that separated compliant platforms from offshore competitors. Trading desks that maintained regulatory relationships captured volume during the transition. Firms without those relationships faced margin compression or exit.

At the same time, the buildout of AI training infrastructure accelerated beyond what existing grid capacity could absorb. Data centers that historically took 18 months to energize now face three to six year timelines for new power connections. Sites with existing utility agreements and completed interconnection studies became immediately differentiated assets. The constraint was not land or capital. It was kilowatts already approved and transformer capacity already installed.

Galaxy holds exposure to both environments. The crypto side consists of trading desks, lending operations, and asset management platforms serving institutional counterparties. The infrastructure side consists of a former mining facility in West Texas that completed large load interconnection studies with ERCOT and secured long-term contracts with CoreWeave for 800 megawatts of AI compute hosting. The first phase begins delivering power in early 2026. Additional studies covering another 830 megawatts received approval in January 2026, bringing total contracted and approved capacity above 1.6 gigawatts.

The timing matters because both constraints tightened during the same window. Crypto trading margins expanded as platforms with proper licensing absorbed share from competitors facing enforcement. Data center power became economically valuable before projects showed up in EBITDA because buyers now pay premiums for certainty of delivery. A site that can energize in 2026 carries different economics than one facing 2028 or 2030 timelines with utility queues stretching into uncertainty.

The crypto business operates on volume sensitivity with minimal carrying costs. When trading activity increases, Galaxy processes more transactions across spot markets, derivatives, and over-the-counter desks. The model scales quickly because the infrastructure already exists. Balance sheets get deployed, collateral gets rehypothecated, and fee income compounds as long as participation continues. Loan books expand when borrowers want leverage. Advisory fees materialize when institutions need counterparties for large block trades or structured products.

This business does not require a perfect environment. It requires activity. Volatility drives trading. Leverage demand drives lending. Institutional allocations drive custody and asset management fees. The revenue shows up within the same quarter that conditions improve because the transmission mechanism involves behavior changes, not construction timelines.

The data center business operates under entirely different mechanics. CoreWeave signed a 15-year lease agreement for 800 megawatts of critical IT load at the Helios campus. Galaxy expects to generate over one billion dollars in average annual revenue across the contract term based on committed pricing and full capacity utilization. The company financed the retrofit with a 1.4 billion dollar project facility secured at 80 percent loan-to-cost, contributing 350 million in equity. Construction timelines determine revenue recognition, not market sentiment.

The revenue structure reflects the economics of scarcity. Data center operators pay for secured power capacity because alternatives involve multi-year delays and regulatory uncertainty. ERCOT interconnection studies now take 12 to 24 months. Environmental permitting adds another layer. Transmission upgrades require coordination across utilities, grid operators, and state regulators. A site that already completed these processes and holds executed service agreements with the transmission provider trades at a premium to undeveloped land with speculative power access.

Galaxy purchased the Helios site from Argo Blockchain in December 2022 for approximately 65 million dollars when it functioned as a bitcoin mining facility. The same site now anchors contracts worth 15 billion over 15 years because the use case shifted from speculative mining to contracted infrastructure hosting. The power did not change. The scarcity premium changed. AI training clusters need continuous uptime and cannot tolerate interruptions. Bitcoin mining can pause during high electricity prices. That difference in operational requirements transformed stranded mining capacity into premium data center real estate.

The capital structure separates the businesses. The crypto operations generate distributable cash flow. The data center business generates contracted revenue but requires capital expenditures during buildout phases. The debt facility funds construction without diluting the parent company during the period when revenue has not yet begun. Once capacity delivers, the contracted payments cover debt service and produce incremental cash flow. The structure allows both sides to scale independently without competing for the same capital pool.

III. The Mispricing or Tension

The market initially priced Galaxy as a crypto-exposed financial services firm with an ancillary infrastructure project. The stock traded with correlation to bitcoin prices and crypto market sentiment because that represented the majority of historical revenue. Analyst models treated the Helios project as a future optionality rather than a secured cash flow stream with contracted counterparties and completed interconnection approvals.

That framing misses the structural shift. The crypto business remains cash-generative and responsive to activity cycles, but it no longer represents the only source of value creation. The data center contracts lock in revenue at pricing that reflects power scarcity rather than mining economics. A facility generating one billion annually in infrastructure revenue trades at different multiples than a crypto trading desk with the same nominal earnings because the risk profiles diverge entirely.

The narrative tension emerges from treating these businesses as dependent rather than orthogonal. When crypto sentiment weakens, observers assume Galaxy's entire model compresses. When data center timelines extend, observers question execution risk. Neither conclusion accounts for the fact that these revenue streams do not need to peak simultaneously to create value. The crypto side produces immediate cash during periods of elevated trading activity. The data center side produces contracted payments regardless of whether bitcoin trades at 90,000 or 110,000.

The pricing also reflects uncertainty about operational delivery. Retrofitting a mining facility into an AI data center involves transforming electrical infrastructure, cooling systems, and network connectivity. The first phase delivers in early 2026, which means the market will soon observe whether contracted revenue materializes as modeled. Until then, the data center business trades at a discount to comparable pure-play infrastructure because Galaxy carries additional volatility from the crypto operations.

Another source of tension involves regulatory and political risk surrounding data centers. Power consumption for AI workloads has become a contentious issue in multiple states, with opposition emerging from both progressive and conservative policymakers concerned about grid strain and electricity costs. ERCOT approved Galaxy's expansion, but future capacity beyond the current 1.6 gigawatts depends on additional load studies and utility coordination that could face delays or opposition. The Helios site benefits from being located in Texas, where grid independence and economic development incentives favor large industrial projects, but broader political backlash against data center power consumption could still constrain expansion timelines or increase permitting costs.

The mispricing exists because the market has not yet fully separated the cash-generative trading business from the contracted infrastructure business. The stock price moved with crypto sentiment even as the company secured long-term agreements that insulate a growing portion of revenue from market cycles. Once the first phase energizes and revenue begins appearing in financial statements, the valuation framework will need to incorporate both the reactive trading business and the contracted infrastructure business as separate drivers rather than conflating them into a single crypto-correlated multiple.

IV. Second-Order Implications

The first-order effect involves Galaxy capturing fees during crypto trading cycles while simultaneously building contracted revenue from data center operations. The second-order effect involves the optionality embedded in controlling additional capacity at a site where expansion already received regulatory approval.

Galaxy's total approved power capacity at Helios now exceeds 1.6 gigawatts, with an additional 2.7 gigawatts under various stages of load study. At full buildout, the campus could support up to 3.5 gigawatts. Each incremental phase that receives approval and secures a tenant creates contracted revenue without requiring the crypto business to perform. The company can layer additional capacity onto the existing site, leveraging infrastructure already in place and relationships already established with ERCOT and utility providers.

This matters because data center power scarcity is not resolving quickly. New generation capacity takes three to six years for solar or wind projects and six years or more for natural gas combined-cycle plants. Nuclear projects extend even further into the future. Meanwhile, demand from AI training and inference workloads continues accelerating. The gap between energization timelines and compute demand creates a window where sites with secured power trade at sustained premiums. Galaxy positioned itself inside that window by completing interconnection studies and securing utility agreements before broader market recognition of the constraint.

The crypto business provides near-term cash flow that helps fund expansion or return capital during periods of elevated activity. Trading revenue in the third quarter of 2025 surged to record levels, with the Global Markets business driving a 1500 percent sequential increase in net income. Loan book size expanded to 1.8 billion dollars. Those results demonstrate that the crypto operations can generate significant cash flow when conditions align, even as the data center business builds toward contracted revenue that does not depend on those same conditions.

The strategic flexibility comes from having two businesses that respond to different catalysts on different timelines. If crypto trading volumes remain elevated, Galaxy captures fees and expands its balance sheet capacity for lending and structured products. If trading volumes decline, the data center contracts continue producing revenue based on delivered capacity rather than market sentiment. The company does not need both businesses firing simultaneously to create value. It needs the crypto business to generate cash during favorable windows while the data center business builds durable contracted revenue that compounds over a 15-year term.

Another implication involves the potential for Galaxy to replicate the model at other sites. The company has stated it is evaluating additional power and land opportunities in Texas and beyond. If the Helios execution proceeds as contracted, Galaxy could use the same playbook at other locations where power access exists but data center development has not yet materialized. That would transform the company from operating a single large campus into running a multi-campus infrastructure platform. The crypto business would continue funding operations while the data center portfolio scales across multiple sites with staggered delivery timelines.

V. Constraints and Limits

The thesis depends on operational execution. Galaxy must retrofit the Helios facility on schedule and deliver power to CoreWeave as contracted. Construction delays, equipment shortages, or permitting issues could push revenue recognition into later periods or increase capital expenditures beyond the financed amount. Data center construction has faced supply chain constraints across transformers, cooling equipment, and backup power systems. Any extension of delivery timelines would delay contracted revenue and potentially trigger renegotiations with the tenant.

Counterparty risk also matters. CoreWeave committed to the full 800 megawatts under long-term agreements, but the company's financial performance depends on securing its own customers for GPU clusters and AI compute services. If CoreWeave faces demand weakness or competition from other cloud providers, its ability to utilize the full contracted capacity could come under pressure. While the lease agreements likely contain take-or-pay provisions, any deterioration in CoreWeave's business would create uncertainty around contract renewal or expansion.

The crypto business remains exposed to regulatory changes and market cycles. Trading volumes spiked in the third quarter of 2025 but subsequently moderated as bitcoin prices consolidated. If crypto markets enter an extended downturn with declining volatility and reduced institutional participation, Galaxy's trading and lending operations would generate lower fee income. The company has built relationships with institutional clients and maintains regulatory compliance, but that positioning does not eliminate exposure to industry-wide volume declines during bear markets.

Political and regulatory opposition to data center power consumption could limit future expansion. While Galaxy secured approval for 1.6 gigawatts and continues pursuing additional studies for another 2.7 gigawatts, the broader backlash against data center energy usage has intensified. Bernie Sanders called for a national moratorium on data center construction. State regulators in multiple jurisdictions have expressed concern about grid reliability and electricity cost pass-throughs to residential customers. If opposition crystallizes into restrictive policies at the state or federal level, Galaxy's ability to expand beyond current committed capacity could face significant delays or denial.

The Helios site benefits from being located in Texas, where grid independence and pro-business policies favor large industrial projects, but even ERCOT faces capacity constraints as demand growth outpaces generation additions. The grid operator projected a six-gigawatt shortfall by 2027, which could trigger emergency measures or price spikes that affect operating economics even for sites with contracted power agreements.

The valuation also depends on how markets separate the crypto business from the infrastructure business. If Galaxy continues trading at multiples driven primarily by bitcoin price correlation, the contracted data center revenue may not receive full credit until it appears consistently in financial results. This creates timing risk where the stock remains volatile around crypto market moves even as the business model diversifies. Conversely, if the market begins valuing Galaxy primarily as an infrastructure company, the reactive cash flow generation from crypto operations could get discounted as non-core.

Capital allocation decisions will influence long-term returns. Galaxy must balance funding data center expansion, returning capital to shareholders, and maintaining balance sheet capacity for crypto trading and lending operations. The company raised 1.4 billion in project financing for the first phase at Helios, but future expansion phases will require additional capital. If the company pursues aggressive buildout across multiple sites, equity dilution or increased leverage could offset the benefits of contracted revenue growth.

VI. Synthesis

Galaxy holds a position where two types of scarcity intersect without requiring synchronization. The crypto business scales immediately when trading activity increases, producing fees from volatility and institutional participation. The data center business produces contracted revenue from power capacity that takes years to replicate and cannot be accelerated once demand materializes.

The setup works because these constraints operate independently. Crypto trading depends on participation, volatility, and institutional demand for leverage and execution services. Data center power depends on completed interconnection studies, utility service agreements, and regulatory approvals that take three to six years for new sites. Galaxy controls both exposures at a moment when each constraint has tightened.

The market has not fully separated these businesses in valuation terms. The stock trades with correlation to bitcoin prices despite holding long-term contracts that deliver revenue regardless of crypto market conditions. That separation will occur as the first phase at Helios energizes and contracted payments begin appearing in financial statements. Until then, the price reflects crypto volatility rather than infrastructure value.

The structural advantage lies in owning capacity where replication timelines extend beyond the window in which buyers need certainty. Data centers cannot wait three to six years for new power generation. They need sites that can energize in 2026 or 2027. Galaxy holds one of those sites with contracted capacity already committed to a creditworthy tenant. The crypto business funds operations while that value gets built. The data center business produces contracted revenue while crypto markets cycle.

The question is not whether both businesses need to peak simultaneously. The question is whether markets will eventually price the cash-generative trading operations separately from the contracted infrastructure cash flows. One responds to activity cycles. The other builds value through capacity delivery in an environment where alternatives involve delays measured in years rather than months.

This report is provided for informational and educational purposes only. It does not constitute financial advice, investment recommendations, or an offer to buy or sell any security. Readers should conduct their own research and consult with qualified professionals before making any investment decisions.

—

This analysis is for educational purposes. It does not constitute investment advice or a recommendation to buy or sell any security. Investors should conduct their own due diligence and consult financial advisors.