Hey! Thanks for being here for Edition #3. You're one of 100+ subscribers, and I genuinely appreciate every single one of you.

Also new this week: I've added a Deal Spotlight section below, where I'll walk through specific deals I'm underwriting - the risk, the location, the downside protection, and the exit. Let me know what you think.

There’s also something I've been sitting with this week…

For years, I replied to every message within minutes - every email, every WhatsApp, every LinkedIn DM. I thought that was professionalism. It wasn't. It was anxiety.

The moment I started protecting my time and switching off, nothing fell apart. Deals didn't stop. The right people still found me. Being "always on" doesn't signal professionalism - it signals you have nothing better to do.

Writing this one from Ho Chi Minh City today. If you're based here, send me a DM - egg coffee's on me.

Let's get into it.

The Philippines formally joined Pax Silica in April - the U.S.-led coalition building secure supply chains for semiconductors, critical minerals, and AI infrastructure outside China. Washington and Manila have paired that with a 4,000-acre industrial zone at the old Clark Air Base, billed as the first "AI-native industrial acceleration zone" under the framework. Technical site assessments are underway now, an initial State Department agreement is targeted for later this year, and Philippine officials say the zone could pull in at least $10B in first-phase investment.

For investors, the interesting exposure isn't AI software - that trade is crowded. It's the physical layer underneath: power, data centres, fibre, industrial real estate, and critical minerals processing. Large anchor tenants tend to reprice land and infrastructure well before the rest of the ecosystem shows up. The open risk is that Manila has already rejected U.S. proposals for special legal carve-outs inside the zone - and that tension will resurface in every long-term lease negotiation from here.

Two more worth flagging from this week:

  • MGX (Abu Dhabi, chaired by Sheikh Tahnoon bin Zayed Al Nahyan, backed by Mubadala and G42) closed one of the largest AI infrastructure war chests on earth - $50B, $500M minimum ticket - and is reportedly circling Singapore's DayOne in what would be Asia's largest data centre acquisition.

  • On Singapore's doorstep, the Johor-Singapore SEZ has already pulled in $15B of approved investment in nine months, with ByteDance, Microsoft, and DayOne committed and 50+ data centres now sitting in Johor.

What's crossing my desk and my conversations this week:

  • A private credit lender financing data centre equipment closed the year with $3B in commitments. Most developers think about DC financing at the real estate level - this lender starts with what's inside: switchgear, transformers, generators, cooling. Their first question on every deal isn't about the asset; it's "who buys the equipment if everything goes wrong?" Tickets up to $500M, backed by a $50B asset manager, will finance both new and used equipment alongside existing bank groups.

  • A UK-based Family Office lending £10M-£200M+ on real estate, securities, aviation, marine, art, and collectibles across the USA, UAE, Central Europe, and the UK builds its entire process around one question: "what's the exit?" They want multiple exit paths before they'll even review a deal, and - unusually - they lend against what you paid, not what the asset is worth. Conservative by design, not accident. They can close in 30 days or less.

  • Data centres are still the dominant conversation across nearly every lender I speak to right now. That's why I'm still building a curated shortlist of the private credit firms actively funding them - who they are, what they need to see, and how to get in front of them. More on that soon.

Two deals I underwrote and ran through my usual four questions - stage of risk, why the location works, how the downside is protected, and whether we control the exit:

A $127M luxury resort in Mexico, built with no bank debt. Stage 1 is done: 43 residences sold at ~$1M each, 30 hotel rooms running at 50% occupancy, $3.95M in current annual revenue. The $127M raise funds Stage 2 - 715 acres of expansion land bought at $10/m² against comps of $200/m². Downside protection is solid ($301M in assets against $81M in debt, 3.7x coverage), and the intended exit is a strategic sale to a name like 1Hotels, Auberge, or Six Senses. This is a strong fit for family offices and equity investors, a medium fit for private credit - 50% occupancy is respectable for a site still under construction, but most lenders want 70-75%.

A $15M green hydrogen plant in Sarawak, Malaysia. Land, permits, and EPC contract are locked in, and the electrolyser (Plug Power) is on order - powered entirely by Sarawak hydropower. The plant pencils out at $9/kg production cost against Japan/Korea buyers currently paying $16-23/kg. My honest red flag: there isn't a single signed offtake agreement yet. No LOI, no contract. Until that changes, this is a strong fit for private credit and family offices comfortable with pre-revenue risk and a weak fit for institutional capital.

I spoke to a PE firm rolling up roofing companies this week. There are 100,000+ roofing contractors in the U.S. - less than 1% will ever see a PE offer. I asked what puts you in that 1%:

  • Commercial work, not just residential insurance jobs

  • Maintenance contracts, not one-off replacements

  • Owner off the tools and out of the sales process

  • Customer concentration below 20%

  • Books that hold up under audit

A single roofing shop sells for 3-5x EBITDA. A professionally managed platform with recurring revenue sells for 6-8x. Home Depot's $18.3B acquisition of SRS Distribution in 2024 shows where that gap can lead.

Most owners think they're selling a truck and a crew. The ones getting rich are selling a repeatable system - and that isn't unique to roofing.

A snapshot of what's currently moving through Capital Arbitrage:

Note: For the best reading experience, open this on a desktop.

Deal

Asset Type

Geography

Structure

Size

EU Workforce Housing

Brownfield

Netherlands

Debt/Equity

400M

UK Data Centre

Brownfield 

UK

Debt

65M

Miami Luxury Residential

Greenfield

USA

Equity

40M

Biomass

Greenfield

USA

Debt/Equity

865M

Landbanking

Landbanking

Hong Kong

Debt/Equity

40M

This isn't the full pipeline - just a flavour of what's active. If you're an investor or lender active in any of these asset classes or geographies, reply to this email.

That's all for this week. See you next Friday.

— Jordon

P.S. I get 500,000+ monthly impressions on LinkedIn and a growing list of private capital readers right here. If you'd like to get your company, fund, or raise in front of this audience, just reply to this email.

Keep Reading