Capital Stack Parallax

Here's a number that should change how you think about the next decade: $150 trillion.

That's the size of the global debt and credit markets. The venture capital industry, by comparison, is roughly $2 trillion. The entire VC industry fits inside the debt markets 75 times over. I'll say that again because it's genuinely wild: seventy-five times. And for the last twenty years, tech founders have built their entire world inside that sliver. They've gotten extraordinarily good at raising equity, telling growth stories, and converting pitch decks into cheques.

It didn't matter. Software was cheap to deploy. You could scale a SaaS company to $100M in revenue on a few hundred million in venture capital because bits are nearly free to reproduce.

That era is ending. Not because software is dying, but because AI is making it radically easier to build. When any competent team can ship a product in weeks that used to take years, software moats are thinner than they've ever been. And the defensibility of a pure software startup has never been more at risk. VCs know this. They're not all keen to fund another SaaS company only to watch three AI-native competitors appear before the Series A clears.

Capital is rotating toward things that are harder to replicate: physical infrastructure, hardware, atoms. The pattern that defined the last decade, vertical SaaS eating every industry with software, is being replaced by something heavier. Vertical robotics. Vertical energy. Vertical infrastructure. Every industry that was digitised by software is now being physicalised by smarter hardware. It's robots in warehouses, batteries in homes, AI data centres that cost $10 billion before they serve a single query. Global capex is exploding. 

This is the hard asset supercycle. And it will not be funded solely by venture capital. The numbers are simply too large. When Crusoe Energy needs $11.6 billion to build a single AI data centre campus, a Series C doesn't cover it. When a robotics company needs to manufacture and deploy ten thousand units to fulfil its order book, equity alone can't carry the load. And shouldn’t either.

But here's the collision: those companies still need equity to get there. You can't walk into a debt market on day one. You need venture capital to prove the technology, find product-market fit, to seed the debt and build the early track record that makes you financeable. Equity lights the fuse. Debt is the rocket fuel. The problem is that most founders only know how to light fuses.

The founders who understand this are already building differently. The ones who don't are going to hit a wall that no amount of vision or charisma can break through.

This piece is about how to avoid that wall. It's about a concept I’m calling Capital Stack Parallax: the ability to present your company simultaneously to two completely different audiences, each of whom needs to see a different version of the same truth. Master it and you unlock the largest capital markets on earth. Fail to master it and you'll join the long list of hardtech companies that had the technology, had the customers, and still couldn't scale.

Buckle in. This one's long. But if you're building anything that touches physical assets, this might be one of those butterfly effect reads that nudges your startup onto the unicorn timeline.

Here's something nobody tells you when you start a hardtech company: to succeed, you need to develop a split personality. You need to exist in two places at once. Like a particle in superposition, your state changes depending on who's observing you.

To VC’s, you're a visionary founder rewriting the rules of a trillion-pound industry. You're disrupting. Risk? That's the whole point.

To debt providers, you're a disciplined operator running a predictable widget. You've got tangible assets, contracted cash flows, and unit economics you could set your watch to.

Same founder. Same company. Two completely different modes.

If that sounds schizophrenic, congratulations. You're beginning to understand how hardtech actually scales.

Most founders only master one personality. They raise equity brilliantly but never unlock debt. Three years in, they need $100M to scale, but VCs are tapped out and banks won't touch them. They've built a science project, not a business.

The ones who win? They practise switching between personas like method actors. Tuesday they're selling the impossible dream. Thursday they're proving they're boring and predictable.

In hardtech, unlike Gotham, being two-faced isn't a character flaw. It's a survival skill.

So founders, let me show you how to channel your inner Daniel Day-Lewis.

OK. Before we go further, I need you to actually understand the capital stack (we’ve developed a Capital Stack Manifesto - read it to understand just how important this is). Not the hand-wavy "equity is expensive, debt is cheap" version you picked up from Twitter. The real thing. I promise this section is worth the effort. Everything else in this piece builds on it.

A capital stack is the complete architecture of how a company is funded, organised from top to bottom by who gets paid last and therefore who takes the most risk. Every layer has a different cost, a different risk appetite, and a different claim on your cash flows. Understanding this isn't optional in hardtech. It's the difference between a company that scales and one that stalls.

Equity sits at the bottom. First money in, last money out. If the company fails, equity holders get whatever is left after everyone else has been paid, which is often nothing. That's why equity demands the highest returns and why it's the most expensive form of capital you will ever raise. Not because of interest rates, but because of dilution. A $50M Series B at a $200M valuation hands over 20% of your company. And that dilution compounds: every subsequent round stacks on top of it. In software, the margins are high enough to absorb this. In hardtech, where margins are thinner and deployment costs are real, it can be fatal.

There's also a behavioural problem with equity that nobody likes to talk about. VCs want to back winners. If they think you're that, they will fight to give you money. A $100M Series B when you needed $40M doesn't just dilute you unnecessarily. It breeds the kind of operational discipline you're more likely to see on a bodybuilder's cheat day. This does not play well with debt markets.

Debt sits at the top. Lowest risk, lowest return, first claim on assets if anything goes wrong. But here's where most founders' understanding stops, and where the real leverage begins. Because "debt" isn't one thing. It's two fundamentally different structures, and confusing them is one of the most common mistakes in hardtech.

On-balance sheet debt is the version most founders encounter first. Venture debt, equipment financing, bank loans. You borrow against your company. The lender looks at your balance sheet, your equity raised, your burn rate, and decides how much they're willing to extend. If you've raised $10M in equity, your on-balance sheet debt capacity is roughly $5M, sometimes less. This is useful. It extends runway. It can fund specific purchases. But it's fundamentally limited by what you've already raised.

Off-balance sheet debt is where the real transformation happens. This is the bit that changes everything, so pay attention.

Instead of borrowing against your company, you create a separate vehicle, typically an SPV, and transfer your performing assets into it: customer contracts, equipment on lease, receivables generating cash flow. The SPV borrows against those assets directly. That debt doesn't sit on your company's books. It sits on the SPV's.

Why does this matter? Because your borrowing capacity is less tethered to what you've raised in equity. It's tied to the quality and volume of the assets you originate and your ability to sponsor that. You won't get dollar-for-dollar. Lenders advance a percentage against those assets. But the ceiling is suddenly far higher than your balance sheet alone would allow. An easy way to think about this is like a mortgage, each asset you’re buying needs a deposit down, just like your house. You're pledging performing assets in exchange for cheaper capital, priced against the cash flows, not entirely against your startup risk.

This is the mechanism that unlocks everything else in this article. Capital recycling, the marriage advantage, the structural leverage that turns $50M of equity into hundreds of millions of deployment. It all flows from the ability to move assets off your balance sheet and finance them independently. It's also how you lighten your own balance sheet to remain attractive to VCs.

There's a related distinction that trips people up constantly: debt and credit are not the same thing in this context. This one is important, because most founders get it backwards.

I'm not talking about using debt to buy a new 3D printer for your lab. That's table stakes. Most founders I meet think too much in the world of inputs: "I need a factory," "I need equipment," "I need to build the thing." This isn't wrong, but it's also not how off-balance sheet debt providers think. They're not in the business of funding your shopping list. They're in the business of buying your cash flows. The lovely team at the premier credit fund you're courting is in the business of money, not of collecting your scrap parts when you fall into bankruptcy. Cash is the protection they need. So shaping your ask around what cash you're generating, not what kit you need to buy, is everything. Collateral quality matters, obviously. But off-balance sheet, it's a non-starter without the cash coming in.

Debt funds your inputs. You borrow to buy equipment, build inventory, acquire the things you need to operate. Credit funds your outputs. You finance the contracts and revenue streams your business generates. A robotics company borrows debt to buy the robots. Then it finances the customer leases through a credit facility. Debt on the way in. Credit on the way out. Two different instruments, two different lenders, two different risk profiles. If you conflate them, you'll either over-leverage your balance sheet or leave massive scale on the table.

Here's something you may of picked up from Linkedin announcements of enormous funding rounds: your credit facility will almost always be the biggest funding number in your company. Often by a lot. Which means your credit line is the thing you should be designing your entire business around. Structure your contracts to be financeable. Standardise your terms. Make your cash flows legible. Because that's where the real scale lives.

Let's be clear about what this means. The moment you start financing assets through an SPV and drawing on credit facilities, your hardware company has metamorphosed into a lender. You are originating financial products. And when you lend, you have obligations to those you borrow from. Serious ones. Reporting covenants. Performance triggers. Servicer standards. If your assets underperform, if your data is late, if your collections slip, you don't get a stern email. You get a default notice. This is not a side project bolted onto your core business. It becomes your core business. The sooner founders internalise that, the better. Because the root of all hardware success at scale is financial engineering + engineering. 

Everything up to this point is asset-backed lending: private facilities between you and your lenders. At the far end of the stack sit asset-backed securities: the endgame. The difference is the public bit. This is when your activity is so proven that you can package contracts into tradeable securities and sell them as listed securities in the publicly traded capital markets  Mortgages work this way. Auto loans work this way. SolarCity did it first for distributed solar in 2013. Enpal did it first for European residential solar a decade later. It's the point at which the most conservative capital in the world is willing to buy your cash flows off the shelf. The equivalent of getting a Michelin star, except the restaurant is your receivables book. Something that most of the startups reading this are far away from, and comes with terms that are representative of the history and quality of the loans these companies are originating. You can not skip to this step yourself.

Senior debt, in all its forms, is cheaper than equity precisely because it's what most VCs would call boring. And boring, in debt capital markets, is a compliment.

Here's the key insight most founders miss: you're building one company, but you're crafting two different investment cases, one for each end of this stack. And if you only remember one line from this entire piece, make it this one: equity's primary role in a hardtech business is not to fund growth. It's to accelerate you toward a capital structure where debt does the heavy lifting. Every dollar of equity you raise should be working toward the day it's no longer the main event.

Let me make this concrete.

The equity universe: Risk is exciting. 100x growth is the goal. You're selling a future that doesn't exist yet. Normal rules don't apply because you're rewriting them. When Crusoe Energy raised its $600M Series D from Founders Fund and NVIDIA at a $2.8 billion valuation, they were selling a vision: the future of clean-energy AI compute. Category creation. A world where data centres are built on stranded energy. That's the equity story.

The debt universe: Risk is expensive. Steady returns are the goal. You're proving you're exactly like every other company that borrowed money and paid it back. Normal rules absolutely apply, and you're grateful for them. When Crusoe secured a $750M credit facility from Brookfield six months later, they were telling a completely different story: long-term compute contracts, physical data centre assets, predictable cash flows, recoverable hardware. That's the debt story.

Same company. Two completely different pitches. Both completely true. It's not spin. It's not deception. It's the same reality described in two different languages for two different risk profiles. Which is, you know, the entire thesis of this article.

The deepest difference between these universes? Their relationship with failure.

VCs expect to lose money on 70% of deals. One 100x return covers everything. Debt providers expect all 30 companies in their portfolio to pay them back. If even two don't, they've had a bad year. Think about that asymmetry for a second. VCs are in the business of structurally encouraging failure. Debt providers are in the business of eliminating it. 

Which is why reaching the threshold of "debt ready" might be the most important signal that you've broken away from being another VC statistic. It means the most sceptical, most conservative capital in the world has looked at your business and said: yes, this is real.

And here's the quiet irony: raising equity is often easier. Not because equity investors are less intelligent, but because the equity process rewards exactly the skills founders already have. Storytelling. Vision. Charisma. The ability to make someone believe in a future that doesn't exist yet. Debt asks you to prove the present. That's a harder conversation, and a less flattering one. Which is precisely why so many founders avoid it until it's too late.

Stage 1: Developing the Activity. You're burning cash to prove the technology works. No debt provider will touch you yet. But even now, start thinking about what they'll eventually want.

What assets are you creating? What cash flows will they generate? What data proves they're predictable?

The founders who nail this stage aren't just building cool technology. They're building financeable activity. And here's the brutal truth: if your historical activity is recorded poorly, it will be worthless for future debt raising. Debt providers don't just want to see that you have activity. They want to see that you can accurately measure and report on it. Think asset-level tracking with serial numbers, locations, and performance data. Standardised monthly financials. Operational dashboards tied to financial KPIs.

Enpal understood this from day one. Even as a small solar installation startup in Berlin, nobody's idea of a future financial powerhouse, they built their data infrastructure as if they were already a financial institution. Every panel tracked. Every contract standardised. Every customer payment recorded in a format that a lender could underwrite. That obsessive early discipline is what allowed them, years later, to securitise €240 million of solar contracts in Europe's first residential solar ABS, structured by Citi, with Barclays, Bank of America, and Crédit Agricole as joint lead managers. Viktor Wingert, CIO and Co-founder of Enpal, said regarding the securitisation that “This commitment enables us to offer our customers a uniquely flexible and affordable solution that otherwise would not be available in the German market”. 

You don't get there without the groundwork. Start building these systems on day one. Not the day before your first lender meeting. Unless you're also building a time machine.

Stage 2: Matching Capital to Activity. You have a working product and early customers. This is when you start matching different types of capital to different parts of your business. Equity still funds R&D and market expansion. But now you're building relationships with debt providers.

The problem? There's no upside for lenders beyond principal plus interest. And right now you're small. If you want to borrow $10M but your deal will take a lender five times longer to close than the $150M deal that just crossed their desk, don't be surprised when you're left on read.

This is the valley of death for hardtech capital structures. You're too proven for pure venture risk but too small for institutional debt. I've watched brilliant founders get stuck in this no-man's-land for years. The companies that cross this valley do so by making themselves easy to underwrite: clean data, standardised contracts, transparent reporting, and a clear asset-by-asset picture of where the money goes and how it comes back.

Stage 3: Scaling with the Right Capital Mix. Your activity is proven and repeatable. Equity funds strategic growth: new products, new markets, new capabilities. Debt funds deploying proven assets at scale. Every dollar of equity can now be leveraged 3 to 5x with asset-backed debt.

Look at where Enpal is today. They've raised roughly $650M in equity at a $2.4 billion valuation. But that equity is dwarfed by their debt infrastructure: €1.1 billion in warehousing facilities from Barclays, Bank of America, and Crédit Agricole. A further €100M backed by the European Investment Bank. Over €5 billion in total financing commitments, with a target of €10 billion by 2027.

Read that again. €650M of equity supporting €5 billion of deployment. That's not a rounding error. That's a fundamentally different kind of company. 

But founders journey need to kick off this journey somewhere.

Let's be direct about this, because I know what some of you are thinking: "Isn't this just spin?"

No. You need both stories because you need both types of capital, for different purposes. And debt, unlike some corners of VC, is particularly is allergic to spin. You often have to change how your contracts are structured, your operations, and counterparties to be compatible with institutional credit. This is not a layer of paint over the top. Or as I would call it, a landlord special. 

Equity funds the "what if": R&D, market creation, moat building. Debt funds the "what is": manufacturing, inventory, tangible assets deployed in the field.

Your debt story starts with proven unit economics. Asset-level analysis. Cash flow predictability backed by contracts. Risk mitigation. And it ends with recovery scenarios: what happens if everything goes sideways.

Different investors, different return profiles, different stories. The founder who can tell both fluently isn't being dishonest. They're being complete.

But fluency doesn't come free. Learning to speak debt is a real project, not a weekend exercise. It means rewiring how you think about your own business. Moving from narrative to numbers. From vision to verification. 

Most founders have spent years sharpening their equity pitch. Debt deserves the same investment. The good news: once you've built it, it compounds. Every contract you structure cleanly, every asset you can point to, every cash flow you can model with confidence that's fluency building on itself.

Treat it like the long game it is.

The hardest part for most founders is knowing when to turn off the vision machine. And, just as importantly, knowing which channel to change too.

There are lines that will kill you in any room if you don't know your audience. "Growth at all costs" sounds like ambition to a VC. To a debt provider, it sounds like a company that hasn't figured out its unit economics and probably never will. "We're not like other companies" sounds like differentiation to one side of the table. To the other, it sounds like a borrower who thinks the rules don't apply to them. "The market doesn't understand our innovation yet" might buy you patience from an equity investor. A lender hears it as: the market has looked at this and said no.

This tension plays out most clearly with FOAK. First-of-a-kind is catnip for equity. It's the whole pitch. Novel technology, novel application, novel market. That's where the outsized returns live. A VC hears "first of a kind" and leans in. A lender hears it and reaches for the door.

And it's not irrational. Debt doesn't get the upside if your breakthrough works. It gets its money back, plus a margin. That's the deal. So the question a lender is asking isn't "could this be huge?" It's "will this work, reliably, on time, and can I get repaid?" First-of-a-kind doesn't help answer that question. It makes it harder.

I spend a lot of my time telling founders to tone it down. Not because what they're doing isn't genuinely new. Often it is, in some meaningful capacity. But that's not the angle that gets you across the line with debt. What gets you across the line is everything around the innovation that is well understood. Proven supply chains. Established offtake structures. A regulatory path that someone has walked before. A construction methodology that has a track record. The more of the stack you can point to and say "this bit is boring, this bit is known," the more comfortable a lender gets.

Risk, for a debt provider, is not about how far you're pushing the frontier. It's about how well they can understand what you're doing. The less a lender has to take on faith, the more they'll put on paper. 

Put nicely, debt funds are a bit more grounded in reality than VCs. Happy to wait for a bit. They see gigs in stadiums. They don't go to open mic nights. They want a guaranteed quality of product, not a chance of one.

There's a subtler point in this also. It's not just about what you say. It's about understanding what product you're actually selling to each side.

When you pitch equity, you're selling shares in a future. The product is upside. The buyer is someone who wants asymmetric returns and is willing to lose everything for a shot at 100x.

When you pitch debt, you're selling a contract. The product is a measurable amount of risk. The buyer is someone who wants their money back, plus interest, on a predictable schedule. They have zero exposure to your upside. They have significant exposure to your downside.

And founders don't love talking about the downside. How the business might fail. How you'd manage a bankruptcy scenario. What the recovery looks like if everything goes wrong. It can feel like being held back, like someone asking you to plan your own funeral. But this is exactly what lenders need to hear. Not because they expect you to fail, but because they need to know you've thought about it with the same rigour they would.

These are not the same product. At all. And yet founders walk into both rooms with essentially the same pitch, then wonder why one of them doesn't go well.

Most hardtech founders have spent their entire careers in the equity world. They've been trained to sell vision, disruption, exponential thinking. That muscle memory is deep. And when they walk into a lender meeting, they can feel that something's off, but they can't quite articulate why. The dance feels wrong. They're doing the cha-cha at a waltz.

I've seen this up close, talking to dozens of venture-backed founders about debt for the first time. There's a moment, almost always, where the energy in the conversation drops. In the equity world, fundraising has a certain romance to it. You're at a coffee meeting. You're riffing on the vision. There's chemistry, spontaneity, mutual excitement. The best founders are phenomenal at this. They're gifted conversationalists who can make a VC fall in love with a future that doesn't exist yet.

Debt doesn't work like that. The starting point of a debt conversation is inherently less engaging, less off-the-cuff, less fun. Nobody falls in love over a data tape review. There's no equivalent of the "what if we could change the world" moment. Instead, there's a spreadsheet (unless you’re with Tangible), a set of covenants, a recovery analysis, and a very specific long list of questions: what are the assets?, what do they earn?, and what happens if you default? It's clinical. And for founders who've built their entire identity around the ability to inspire, it can feel like being asked to do their job with one arm tied behind their back.

The vast majority of hardware startups give away far more of their company than they needed to. Many of those had working technology and real demand. What they didn't have was a capital structure that could fund deployment without reaching for the equity round every time. Each raise that could have been debt was another chunk of the cap table gone.

This is the argument Packy McCormick and I made in Capital Intensity Isn't Bad. Crusoe Energy's Chase Lochmiller put it plainly: "there are a lot of ways founders can end up owning a larger percentage of the company by finding the right investor for the right component of their overall capital stack." Crusoe built 86 data centers and is now overseeing Stargate — pairing equity with asset-backed facilities from the start so physical assets never had to come out of the cap table. Capital intensity and dilution are not the same thing. It's not about how much capital you need. It's about where it comes from.

And there's one more thing founders consistently underestimate: the sheer complexity of structured debt. This isn't like negotiating a valuation over coffee. A structured debt facility can run to thousands of pages of legal documentation. Covenants, triggers, reporting obligations, waterfall mechanics, events of default. Knowing even how to frame what you’re asking for can have you in a flap. Every one of those clauses is a rule your business has to live inside for the life of the facility.

When a founder tells me about the bizarrely cheap debt she's just secured, my first reaction, as she smiles, is a nervous smile back. Because cheap debt often comes with tight constraints. And the question isn't what the interest rate is. The question is what happens when you trip a covenant, when you miss a reporting deadline, when you need to pivot your product and discover that your facility agreement doesn't allow it. Debt can be both a cheaper form of capital compared to equity but also the more restrictive of the two if you get the terms wrong.

This is why working with people who've actually structured these deals matters. Not just lawyers, though you'll need those. Advisors who've been part of raising billions in structured facilities and who understand how a single clause can strangle a business or set it free. The difference between good debt and bad debt isn't the rate. It's the five hundred pages of terms you need to pay attention too.

One of the most common things I see is CEOs treating debt as something that can be delegated down. And look, handing it to a CFO makes sense. That's what CFOs are for. But here's the catch: most startup CFOs have never structured an asset-backed facility in their lives. They've managed burn, run payroll, maybe closed a venture debt round. Off-balance sheet financing is a different discipline entirely. So if you know this is where you're heading, hire for it specifically. Don't assume the CFO title comes with the experience. And if it ends up sitting with a junior hire while the CEO focuses elsewhere, that's somewhat understandable, you're busy. But it's a mistake.

To talk openly to you all. And in an attempt to hand out some helpful truths to the industry as a whole. I think this is often driven by insecurity, not capacity. Equity fundraising is a game most founders have played for years. They know the room, they know the script, they know how to perform. Debt is unfamiliar territory. Different language, different instincts, different questions. And when founders don't feel fluent, they don't lean in. They hand it off.

The problem is that debt doesn't work that way, it’s not a bolt on. Credit providers want to sit across from the person who actually runs the business and hear them explain, in detail, how the assets perform. How the contracts are structured. What happens when things go wrong. Outsource that conversation to someone three levels removed from the decisions, and you won't just slow the process down. You'll kill it.

Here's how you can rewire your brain to make this the priority it needs to be: done right, debt will likely become your largest source of capital. Bigger than any equity round you'll ever raise. So why wouldn't you pay it the same attention? The discomfort isn't a reason to delegate it. It's a reason to get prepared. Your critical path as a successful company has many different debt related milestones. So it’s time to make more space on your to-do list.

This isn't just a hardtech phenomenon. The best capital structure operators in any capital intensive businesses have already figured this out. Just take a look at all the fintech lending unicorns.

Take Ramp. To equity investors, they're the fastest-growing spend management platform in history. $2.3 billion raised in equity. Valuation surging from $7.65 billion to $32 billion in eighteen months. Founders Fund, Khosla, General Catalyst, Lightspeed. The pitch: replacing legacy corporate finance infrastructure with AI-native software. A billion dollars in annualised revenue. 50,000 customers including Shopify, Anduril, and Notion.

But that's only half the story. Every time a customer swipes a Ramp card, someone has to fund that transaction. Ramp manages several hundred million dollars in warehouse credit facilities from Goldman Sachs and Citibank. That debt capital is what actually fuels the originations, what keeps the cards working, what turns the software into a business. Goldman doesn't care about Ramp's AI roadmap. They care about the receivables: how quickly do customers pay, what's the default rate, what's the recovery on delinquent accounts? 

The equity funds the platform. The debt funds the product. Strip away either one and the business doesn't work. This is a $32 billion company and most people who write about it only discuss the equity side.

Now take Enpal, and you see the same pattern in physical infrastructure. They didn't just build a solar installation company. They built a solar fintech company. In 2023, they priced Europe's first-ever residential solar asset-backed security: €240 million of solar contracts, structured by Citi, purchased by 18 institutional investors, and publicly traded on the Luxembourg Stock Exchange. The European Investment Bank came in as anchor investor for €50M, with the European Investment Fund guaranteeing another €50M. They understood from day one that solar panels aren't just their product. Solar panels are the hardware layer of a financial product.

These founders are the ones the hardtech companies should be looking to emulate. 

I saved this section for near the end because I think it's the kind of lofty idea you need to build too. 

Every inflection point in physical infrastructure follows the same pattern. The technology gets invented. It works. People want it. But it doesn't transform civilisation until asset-backed finance makes deployment economically viable at scale. This pattern is so consistent across history that once you see it, you start seeing it everywhere.

In 1919, General Motors had a problem. They were building more cars than people could afford to buy. Banks refused to offer consumer auto credit. They thought it was beneath them. So William Durant, GM's founder, wrote a letter to J. Amory Haskell and created the General Motors Acceptance Corporation: GMAC.

The idea was simple: instead of asking customers to pay for a car in full, let them pay over time. GMAC would buy the receivables from dealers, freeing up capital for more production. The car itself was the collateral.

Henry Ford hated it. He thought instalment buying was reckless, undisciplined. He offered customers a lay-away plan instead: save up, then buy.

History chose GMAC. By 1963, they had financed over 43 million new cars. By the end of the century, GMAC had loaned out more than a trillion dollars. Today it's Ally Financial, one of the largest auto lenders in America. Henry Ford, for all his manufacturing genius, got this one completely wrong.

The car didn't just replace the horse. It restructured society. Suburbs. Highway systems. The entire geography of how humans live and work. And that transformation didn't come from Henry Ford's assembly line. It came from GMAC's balance sheet.

The same pattern played out again and again. Mortgages turned houses from things the wealthy owned into assets that ordinary families could finance, creating the middle class and suburban development as we know it. Aircraft leasing turned planes into accessible infrastructure, shrinking the world. Equipment financing turned container shipping into a global system.

All of this CAPEX innovation. And the transformation happened not when the technology was invented, but when finance made deployment viable at scale.

The internet didn't need this. Software didn't need this. Bits are nearly free to reproduce. A SaaS company could scale to $100M ARR on a few hundred million in venture capital because the marginal cost of serving another customer was essentially zero in the context of most SaaS. But ironically now not for AI businesses. 

Atoms are not SaaS, although they’re often paired together. And right now, we're watching the tech industry collide with this reality in real time.

Look at what's happening. The last decade's winners were software companies. Vertical SaaS ate industry after industry because deployment costs were trivial compared to the potential size of the win: spin up a server, onboard a customer, collect recurring revenue. Equity funded the growth, and the margins were extraordinary.

Now look at the companies defining the next decade. They aren't vertical SaaS. They're vertical robotics. In Q1 2025 alone, over $2.26 billion flowed into robotics funding, with vertical-specific companies accounting for more than 70% of that capital. The global robotics market is projected to grow from $74 billion to $185 billion by 2030. Over 20 private robotics companies are now valued above $1 billion. Figure raised $1 billion in a single round. Physical Intelligence raised $400M. Apptronik closed a $935M Series A.

These aren't research projects. These are companies with validated customer demand and physical hardware to deploy. And they all face the same constraint: deployment costs money. Real money. The kind of money that equity alone cannot provide at the pace the market demands.

Beyond robotics, AI infrastructure is consuming capital at a scale the venture industry has never seen. A single data centre campus now costs $10 billion or more. Goldman Sachs forecasts trillions flowing into AI infrastructure, energy transition, and reshoring over the coming decade. Robot installations are accelerating. Battery deployments are compounding. The entire built environment is being rewired.

This is the hard asset supercycle. And the playbook has flipped. Software companies could spin up capacity instantly. The bottleneck was always demand: could you acquire customers fast enough to justify the growth spend? In hardtech, the bottleneck is reversed. Demand often outpaces supply. You can sign contracts faster than you can build and deploy the hardware to fulfil them. The constraint isn't finding customers. It's manufacturing capacity quickly enough to serve them. And capacity costs real money. That's a fundamentally different capital problem, one that equity alone was never designed to solve.

The founders who understand this are already building accordingly. They're raising equity to prove out the unit economics, then unlocking debt to deploy at a pace equity alone could never sustain. Every customer contract becomes a financeable asset. Every deployed robot, every installed battery, every commissioned data centre becomes collateral for the next wave of deployment.

This is the car moment for an entire generation of technology. Not one product, but dozens of categories simultaneously reaching the point where the technology works, the demand is proven, and the only bottleneck is capital deployment.

I think we'll look back at this decade the way we look back at the 1920s. Not as the era when these technologies were invented. As the era when finance made them accessible to everyone.

We're standing at a unique moment.

There's more capital available than ever: $300 trillion in global financial assets looking for returns. There's more breakthrough technology ready to deploy than ever: batteries, robotics, distributed energy, advanced manufacturing, space infrastructure, automated mobility.

The question is whether the founders building these companies can speak the language of the capital that's waiting for them.

Distributed energy creation and storage will reshape civilisation the way car ownership did. Every home becomes a mini power station. Compute moves to the edge. Manufacturing goes local. Robotics enters every workplace. Sustainable transportation becomes the default. Space becomes accessible. Carbon gets reliably captured and stored. Products previously shipped from China are built round the corner.

This generation of founders needs more range than any before it. The new full stack engineer doesn't mean backend and frontend. It means the whole capital stack. It's not enough to be a brilliant engineer or a charismatic storyteller or a sharp operator. You need to be all three, and you need to switch between them depending on who's sitting across the table.

Each of these civilisation shifting technologies will have its GMAC moment. The technology is ready. The capital is ready. What's needed is founders who can exist in two places at once. Who can be observed by a VC and appear as a visionary, then be observed by a lender and appear as a reliable custodian of debt.