Category Archives: Innovation

Sequel: Apollo’s AI–Infra Flywheel, and the policy needles the UK must thread

Posted on by
Reply

TL;DR (the needle): Apollo isn’t just writing cheques—it’s assembling a time-to-power machine: buying a developer (Stream), a cooling/heat-exchange supplier (Kelvion), and an AI solutions integrator (Trace3). That vertical stack compresses delivery timelines where the bottleneck really is (power, permits, thermal), which is exactly what capital wants to finance at scale. noahpinion.blog Apollo+1

What changed since the first post

1) The scale moved from “big” to “macro.”
The FT now frames the AI infra boom at ~$3T by 2029, with single projects scoped at $100B+ (OpenAI “Stargate”, xAI “Colossus”, Meta mega-campuses). This isn’t just colocation growth—it’s nation-scale capex that outstrips hyperscaler cash flows, pulling in private credit, securitizations and ABS at speed.

2) Financing templates are crystallising.
We’re seeing record-sized, multi-tranche packages (e.g., Meta’s ~$29B) and structured leasing pipelines (e.g., Oracle) that term out tenant risk and recycle developer equity faster—useful context for how the Apollo–Stream flywheel monetises time-to-power.

3) This capex is already a GDP lever.
Paul Kedrosky’s work (and interview) ties AI data-centre spend to a meaningful chunk of recent US GDP growth (a conservative ~0.6–0.7pp of a ~3% quarter, with multipliers pushing the contribution higher). It’s a private-sector stimulus—large, fast, and unusually concentrated. Paul Kedrosky

What Apollo is actually building (and why it matters)

  • Developer core (Stream): access to land, interconnect queues and pre-lets, then refinance with IG debt/ABS—capital velocity is the product. noahpinion.blog

  • Thermal moat (Kelvion): heat-exchange & liquid-cooling hardware is now “schedule-critical” (power density + water scrutiny). Owning a piece de-risks delivery and opex vs. waiting on supply chains. ApolloGlobe Newswire

  • Demand catalyst (Trace3): enterprise AI integration that pulls workloads into Apollo-backed capacity, smoothing lease-up/utilisation assumptions. Apollo

Why that stack wins: in 2025, development financing is projected to set another record (≈10GW breaking ground; ≈$170B of assets needing development/permanent financing). Whoever reliably shortens time-to-power captures that spend and the recycling premium. Bloomberg

The US read-through

  • Growth tailwind—then an air pocket? If AI capex slows, the drop could show up visibly in GDP prints. That’s because today’s growth boost is unusually concentrated (and not very jobs-intensive vs. fulfilment centres).

  • Perishable capex, not railways. GPUs turn over on ~3-year cycles; missing utilisation targets forces early write-downs/refresh—unlike century-life fiber/rail.

  • Financing opacity risks. Rapid growth of off-balance-sheet SPVs, leasing and stacked vehicles (REIT exposure included) can obfuscate risk—watch the private-credit plumbing, not just bank balance sheets.

  • Grid friction. Interconnection studies & queue times are the long pole—typical projects built in 2023 spent ~5 years from request to COD. That’s why developer quality and pre-work (permits, substations) price like gold. Kirkland & Ellis

The UK opportunity—if we pull the right needles

The UK has declared intent: ~£1bn to turbocharge national compute (targeting ~20× capacity over five years). But capital—Apollo included—will only show up where time-to-power is predictable and water/power constraints are “bankable.” Datacenter Dynamics

Today’s friction points

  • Power: well-documented West London capacity constraints slowed large connections—exactly the kind of uncertainty that deters developer-finance flywheels. YouTube

  • Water: the government’s own paper calls for mandatory, location-based water-use reporting and better integration of water planning into AI/DC development. That transparency is overdue and investable. GlobeNewswire

Needles for the UK Government (actionable & investable)

  1. Make WUE data investable (not optional).
    Enact mandatory, auditable WUE (Water-Use Effectiveness) disclosure for large DCs, with real-time metering and site-specific reporting. Tie consent conditions to peak-day water draw, not just annual averages. This aligns directly with DSIT’s call for location-based reporting and tech adoption. GlobeNewswire

  2. Fast-track low-water cooling.
    Update planning and Building Regs guidance to prefer closed-loop liquid cooling/direct-to-chip over open evaporative systems where feasible, and require non-potable/recycled sources when available. Pair that with credits for heat re-use (district networks), which Kelvion-type kit makes easier to standardise. GlobeNewswire

  3. De-risk the interconnect.

    • Create a “Compute NSIP” lane (Nationally Significant Infrastructure Project) for AI campuses that meet strict water/heat criteria, granting accelerated DCOs and coordinated grid works with NGESO/DSOs.

    • Allow private-wire/behind-the-meter generation (renewables + storage; CHP/fuel cells) to count toward capacity tests where it genuinely reduces grid draw. Both steps shorten time-to-power, which is what developer finance prices. YouTube

  4. Finance the time, not just the tin.
    Launch a Compute Connections Facility (Treasury + UKIB) offering recoverable advances for substations and shared grid upgrades, repaid via regulated tariffs as capacity comes online. This crowds in private credit for shells/fit-outs while removing the “first-mover penalty” on grid spend. (Think UK fibre’s duct-sharing lesson, applied to electrons.)

  5. Adopt “use-when-green” economics.
    Encourage time-of-use compute: training jobs scheduled to match renewable peaks (via TOU-based network charges and dynamic connection agreements). It cuts curtailment and lowers the perceived water/power footprint—both bankable to lenders.

  6. Publish time-to-power league tables.
    Quarterly by planning authority/DSO: average months from application → NTP → energisation; water approval lead-times; share of recycled/non-potable water; % heat re-use. Regions will compete on the metric that matters to capital.

How Apollo’s stack aligns with the UK needles

  • Developer + supply-chain control = schedule certainty. Stream’s development engine plus Kelvion’s thermal kit reduces the two biggest UK unknowns: grid & water. Marry that with Trace3’s demand-pull and you have a credible pipeline to anchor private credit issuance here—if the regulatory path is clear. noahpinion.blog Apollo+1

  • Global capital is timing-sensitive. With another record year of development financing projected globally, projects will land wherever time-to-power is shortest—Spain/Nordics/US Sun Belt or the UK, depending on these rules. Bloomberg

Risks to watch (and how policy can mute them)

  • Overbuild/obsolescence: three-year GPU cycles make under-utilisation costly; consent conditions that require heat re-use and water-efficiency upgrades on refresh can protect the public interest if economics change.

  • Financing opacity: SPVs/ABS and REIT exposure can mask leverage. Require enhanced disclosure on any publicly supported project (grid advance, land grants) covering financing stack, lease cover ratios, and refresh obligations.

  • Macro “air pocket”: if the AI capex hose slows, growth prints will show it. A shovel-ready grid queue + “use-when-green” tariffs keep the UK shovel-worthy even as global cycles ebb.

Investor take

For allocators, this is a developer-alpha moment. The edge is in compressing time-to-power and de-risking thermal/water. Apollo’s three-pronged move (developer + thermal + demand) is a signal of where value will accrue. For the UK, the prize is growth without a water backlash—won by making time-to-power transparent, financeable, and fast.

Sources & further reading

  • FT: the $3T AI buildout; mega-project scope; private-credit role.

  • JLL: 2025 outlook—record development financing; ≈10GW breaking ground; ≈$170B assets requiring financing. Bloomberg

  • Apollo press releases: Stream majority stake; Kelvion acquisition; Trace3 acquisition. noahpinion.blog Apollo+1

  • Kedrosky: AI capex as GDP lever; off-balance-sheet risks (essay + interview). Paul Kedrosky

  • LBNL: US interconnection queues; ~5-year median from request to COD for projects built in 2023. Kirkland & Ellis

  • UK policy: DSIT compute roadmap & funding; West London capacity constraints; UK water report (mandatory, location-based reporting). Datacenter Dynamics YouTube GlobeNewswire

Apollo buys “time-to-power”

Posted on by
Reply

At a glance: Apollo (NYSE: APO) is buying a majority stake in Stream Data Centers, a developer with a 4+ GW pipeline and near-term campuses in Chicago, Atlanta, and Dallas (roughly 650 MW of power capacity coming through). Translation: Apollo wants to manufacture AI-grade capacity—land + power + permits—then lock in long leases and recycle capital. That’s a different playbook from buying stabilized boxes via REITs. ApolloApollo Global Management, Inc.Barchart.com

The needles

1) The bottleneck is power and permits, not tenants.
Hyperscaler demand isn’t the scarce resource; interconnects, substation timelines, and entitled land are. Developers with utility relationships and shovel-ready sites capture the biggest spread because they move from “paper megawatts” to energized capacity faster than a yield vehicle can. Apollo is effectively paying for Stream’s optionality on power across its multi-GW pipeline—precisely where the industry choke point sits. JLL expects another record year for development financing in 2025 with ~10 GW breaking ground and ~$170B of asset value needing funding; the capital is chasing power, not occupancy. JLL

2) Follow the capital stack: private markets are setting the terms.
Public REITs optimize for steady yield on stabilized assets. Developers optimize for IRR from risk steps (site control → interconnect → pre-let → NTP → COD). Apollo is dropping long-term capital into those risk steps—then can term out with IG-style, asset-backed structures once leases are signed. If you want the template for where this is going, look at Meta’s ~$29B Louisiana financing led by PIMCO (debt) and Blue Owl (equity): single-project, investment-grade scale, private credit sitting where banks used to. Expect developer + private credit pairings to proliferate. ReutersYahoo Finance

3) Why Stream, why now?
The deal formalizes the developer–finance flywheel: secure land and interconnects → pre-lease to a hyperscaler → raise cheap capital against contracted cash flows → recycle into the next campus. Stream’s specific city mix matters: Chicago, Atlanta, Dallas pair large load growth with comparatively tractable permitting and transmission plans versus ultra-constrained hubs. That positioning helps compress time-to-MW, which is the new currency of returns. Apollo’s own release underscores the scale: 4+ GW in the pipeline, “deployment of billions” into U.S. digital infrastructure. ApolloBarchart.com

4) The macro backdrop is insanely capital-hungry.
McKinsey pegs $6.7T of data-center capex needed by 2030 to keep up with compute demand, with AI-capable facilities taking the lion’s share. In that world, “own the development spread” is a rational, repeatable strategy—especially for sponsors who can bring both patient equity and structured financing to each tranche. McKinsey & Company+1

What this really signals

  • Developers become the new “growth REITs.” They’re not collecting coupons; they’re monetizing milestones. That’s where alpha sits until grid constraints ease. Datacenter Dynamics

  • Private credit is graduating to utility-scale socialization of risk. The Meta structure shows that once you have long-dated leases, you can finance a campus like a toll road or pipeline. Expect tighter spreads and more off-balance-sheet funding for hyperscalers. Reuters

  • Public-market investors aren’t shut out. If you don’t have access to Apollo-style funds, watch proxies: select data-center REITs with development pipelines, transmission and switchgear vendors, and utilities accelerating capex plans in Stream-style markets. JLL’s build forecasts give a practical map for where financing will actually clear. JLL

Risks to the bull case (and how to track them)

  • Overbuild / efficiency shock: If AI inference efficiency jumps (model compression, better GPUs), utilization assumptions could slip, pressuring lease rates. Tell: slowing pre-lets or rising concessions in developer disclosures and broker chatter. (Use JLL quarterly notes as a barometer.) JLL

  • Grid delays: Interconnect queues or local moratoria can push COD right. Tell: slippage in targeted energization dates on Chicago/Atlanta/Dallas campuses. (Stream/Apollo updates, utility IR filings.) Apollo

  • Cost of capital whiplash: If IG appetite cools or private credit reprices wider, recycle math compresses. Tell: fewer large single-asset prints after the Meta deal, or materially wider spreads on campus ABS/bonds. Reuters

Investor checklist (practical signals to watch each quarter)

  1. MW under construction vs. energized at Stream-peer developers—are CODs holding? Datacenter Dynamics

  2. Pre-lease coverage and escalators on new campuses (are 2–3% annual bumps sticking?). Datacenter Dynamics

  3. Financing prints: any follow-ons to Meta’s IG-style package (size, tenor, spread). Reuters

  4. Utility capex plans in IL/GA/TX that unlock substation timelines (proxy for Stream’s city mix). JLL

  5. Sponsor behavior: Does Apollo recycle quickly (refi/partial sale) or hold longer? That will telegraph how rich the development spread remains. Apollo

Bottom line: Apollo didn’t just buy data centers; it bought time-to-power and the right to repeatedly monetize the development curve. In an AI buildout that needs trillions and is starved for interconnects, the developer is the fulcrum. If you’re a public-market investor, shadow that play by tracking power-rich markets, development-heavy REITs, and the cadence of private credit deals that term out risk at scale. McKinsey & CompanyJLLReutersApollo