Skip to content
Substrata
Join

Capital

Money is rarely the scarce thing. Money with the right mandate and the right patience is — and for most of the bottlenecks assessed here, funding is judged not to be what is missing.

Kinds of capital
How this is computedKinds = rows in the capital-kinds list; providers = named funders, each linking its own mandate page; assessed = bottlenecks with a funding judgement; "money is not it" = those judged "not-money".Method →Code →
and what each will not fund
Providers
How this is computedKinds = rows in the capital-kinds list; providers = named funders, each linking its own mandate page; assessed = bottlenecks with a funding judgement; "money is not it" = those judged "not-money".Method →Code →
each citing its own mandate document
Bottlenecks assessed
How this is computedKinds = rows in the capital-kinds list; providers = named funders, each linking its own mandate page; assessed = bottlenecks with a funding judgement; "money is not it" = those judged "not-money".Method →Code →
is funding actually the constraint?
Where money is not it
How this is computedKinds = rows in the capital-kinds list; providers = named funders, each linking its own mandate page; assessed = bottlenecks with a funding judgement; "money is not it" = those judged "not-money".Method →Code →
of 10 assessed

This is not investment advice. It maps who funds what, on whose mandate, over what horizon. Nothing here says where to put money and nothing here is a recommendation. If you want the argument rather than the tables, read what money cannot buy.

01Kinds of money, and what each will not fund

The last column is the one that matters
KindWill not fund
Venture capital

Money for young companies, in exchange for a share of them, accepting that most will fail.

A factory. Venture returns depend on a small number of enormous outcomes, and a plant that earns a steady margin on a known product cannot produce one — however badly the world needs the plant.
Growth equity

Money for a company that already works and wants to get bigger faster.

First-of-a-kind plant risk. If nobody has built this exact thing before, the return cannot be underwritten on comparable companies, which is what this kind of capital is good at.
Project finance

A loan secured against one specific asset and the money it will earn, rather than against the company that builds it.

Anything without a bankable contract. This is the largest pool available to heavy industry and it is also the fussiest — no offtake, no loan, whatever the strategic case.
Public markets

Shares and bonds anyone can buy, priced continuously by whoever is trading them.

A decade of losses without a story that survives quarterly scrutiny. The capital is patient only as long as its holders individually are, which is not the same thing.
Corporate balance sheet

A large company spending its own cash on its own capacity.

Capacity for a competitor, or a shared facility that would help the whole industry. It will not build anything it cannot keep.
Sovereign wealth

A state investing a surplus, usually from resources, on behalf of a country.

Anything politically awkward at home. The horizon is the best available anywhere; the constraint is mandate, not time.
Government grant and subsidy

Public money given, not lent, to make something happen that otherwise would not.

Reliably or quickly. It moves on political cycles, arrives with conditions, and can be withdrawn — which is why it is a poor foundation for a fifteen-year build even when it is generous.
Export credit

A state agency lending or guaranteeing, so that a buyer abroad can afford to buy from that country’s exporters.

A project with no domestic content. The money follows the exporter, not the need, which is why it can build a plant on one continent and never on another.
Development bank

A bank owned by governments, lending for things judged to be in the public interest.

Anything outside its mandate, however commercially sound. It is patient and cheap and narrow, in that order.
Offtake and prepayment

A buyer paying in advance, or committing to buy for years, so the producer can raise the money to build.

Anything the buyer could simply purchase on the market. It appears exactly when a buyer becomes frightened about supply, which makes it a useful signal in itself.
What these columns mean
Cheque
The size of commitment this kind of money typically writes at once.
Patience
How long it will wait before it wants its money back. For a fifteen-year asset this is the field that eliminates most candidates.
Offtake
A buyer committing to purchase output for years, which is usually what makes a new plant financeable at all.

02Is funding actually the constraint?

6 of 10 say no
  • Large power transformer slotsMoney is not the constraint

    Capital has been available for two years without shortening the queue. What binds is qualified winding capacity, core steel, and the fact that almost every unit is custom — none of which a cheque shortens.

  • Grid interconnection queuesMoney is not the constraint

    An administrative queue. There is nothing to buy. Money can fund the transmission that shortens it eventually, but the queue itself is a process, and process is relieved by rules.

  • EUV lithography scannersMoney is not the constraint

    One supplier, sold out, and the buyers are the most cash-rich companies on earth. Every marginal dollar is already bidding for the same machines.

  • Semiconductor process engineersMoney is not the constraint

    Money raises salaries, which moves people between employers without creating any. The constraint is years of experience, and there is no market in that.

  • Crucible-grade high-purity quartz sandMoney is part of it

    Funding could develop a second deposit or a synthetic route, and neither is currently attractive at the price the material commands. The geology is fixed; the willingness to pay for an alternative is not.

  • Grain-oriented electrical steel (GOES)Money is part of it

    A new mill is buildable and expensive, and the reason few are built is that demand is thought temporary. That is a capital judgement, so capital could change it — slowly.

  • REBCO superconducting tape, 12 mmMoney would genuinely help

    Production lines are small and the makers are young companies. This is one of the few rows where writing a cheque genuinely adds capacity, which is why fusion programmes have been buying equity in their own suppliers.

  • Dysprosium metalMoney is part of it

    Mining is fundable and spreading. Separation and metal-making are where the concentration is, and those need capital plus a buyer willing to sign a long contract at above the Chinese price.

  • Electronic-grade polysiliconMoney is not the constraint

    A US plant is at risk of closing because a trade measure drove its customers away. Capacity is not short of funding; it is short of buyers.

  • Silicon carbide substrate, 200 mm semi-insulatingMoney is not the constraint

    Capacity was built ahead of demand that did not arrive, and a maker has been through Chapter 11. More money would make the oversupply worse.

These are judgements, dated like every other judgement here. The reasoning is next to each one so it can be argued with.

03Who could fund relief

4 tracked, each from its own mandate

Three, not more. Two others were drafted and removed because their mandate sentences could not be found on their own pages — the same bar the policy section uses. See project finance and offtake for the mechanisms most of this turns on.