Understanding innovation · Part III · Capital and value
09 · The capital of innovation
14 min read
In brief
- Innovation has a structural financing problem, not an accidental one: the knowledge produced spills over beyond whoever pays for it (spillovers), so the market on its own finances less of it than is in society's interest. This is the economic justification for public intervention, from grants to tax incentives.
- There is no such thing as "capital": there are capitals with different profiles of accepted risk, required return, horizon and rights. A well financed innovation path assigns to each stretch the source whose trade is that level of risk.
- Phased financing is the rational structure for uncertainty: capital committed in tranches, conditional on the evidence produced, with each phase reducing the risk for whoever enters afterwards.
- Capital is not only money: validated problems, ideas, technologies, competencies and networks reduce risk or add value exactly as money does, and can be recognised and remunerated as contributions.
- The repayment order in adverse cases is the most honest map of the distribution of risk: it says who loses first, and therefore who is really taking the risk.
Section 1
Why the market on its own finances too little innovation
Economic theory has a precise answer, formulated by Kenneth Arrow (1962), to the question of why the financing of innovation cannot be left entirely to the market. The knowledge produced by research has properties that make it an anomalous good: once created, using it does not consume it, and excluding others from using it is difficult and imperfect. It follows that whoever invests in research does not capture all the value they create: a part spills over towards competitors, suppliers, customers and whole sectors (the spillovers), and the empirical estimates accumulated over decades place the social return on R&D stably well above the private return. The consequence is systematic underinvestment: every single actor, rationally, invests less than would suit the whole.
To this are added the informational asymmetries of innovative projects (whoever proposes knows much more than whoever finances, and the financier struggles to distinguish the good projects), the absence of collateral (the main asset is knowledge, difficult to pledge) and the long horizon. The practical result is the geography of chapter 06: stretches of the path, in particular the valley of death between TRL 4 and 7, where the need is at its maximum and the spontaneous supply at its minimum.
This analysis is the standard economic justification for public intervention in innovation, in its three main forms: direct financing of research (the tradition that Vannevar Bush's 1945 report institutionalised, and whose depth Mazzucato has documented by showing how much "private" technology rests on public research upstream, from GPS to the internet to biotechnology); grants and subsidised loans to development projects; and tax incentives for R&D, which reduce the private cost of the activity while leaving the selection of projects to the market. Tax incentives in particular exist in most advanced economies in different forms (tax credits, enhanced deductions, favourable regimes for income from intellectual property), and their logic is always the same: to bring the private return closer to the social one, so that the R&D that would not otherwise be undertaken is undertaken. The Spanish Tax Lease applied to R&D, with its combination of deductions and dedicated investment structures, belongs to this family: a mechanism that transfers part of the tax benefit to whoever finances the research, making financeable the phases that ordinary capital would not cover.
Section 2
The sources and their division of labour
"Capital" in the singular is an abstraction that hides the decisive variable: every source has a profile (the risk it accepts, the return it requires, its horizon, the rights it demands, the moment at which it can enter), and the financial design of an innovation path consists in assigning to each stretch the source whose trade is that level of risk. The main families, in the typical order of entry:
The promoter's own capital enters where no third party enters: the validation of the problem, ideation, the concept. It is the stretch with no evidence to show, and whoever covers it is buying the option on all the rest.
Tax incentives for R&D typically cover the phase of research and experimental development: their return for the tax investor derives from the tax saving, and therefore does not depend (or depends only in part) on the commercial success of the project. It is the property that makes them suitable for the most uncertain stretch: they remunerate the financing of research as such.
Non-repayable public grants finance the projects that pass a competitive selection on criteria of merit and impact. They are not repaid and they do not touch the share capital: they dilute nobody, and they absorb risk before private capital enters. The calls define the phases they cover, as a rule from experimental development onwards.
Subsidised public loans often accompany grants in the same calls: low or zero interest rates, long maturities, an initial grace period. They do not dilute, but they are debt and they weigh on the balance sheet of whoever receives them: the distinction between the non-repayable part and the part to be repaid is among the first things to read in a call.
Private risk capital (angels, venture capital, convertible instruments) buys a shareholding in the residual uncertainty: its return depends entirely on success, and for this reason it demands the high returns that chapter 10 will quantify. It can enter early or late; what changes with the moment is the price of the risk it buys.
Ordinary debt (pure lender, bank) enters when there are flows or assets that repay it: an agreed return, no participation in the value created, priority in repayment. It is the cheapest source and the least suited to uncertainty: asking for debt for a project in research is an error of assignment, like paying with equity for a phase that debt would cover.
Two principles govern the composition. The first is complementarity: the sources do not compete, they are layered, and well designed combinations reinforce one another (private capital added to a project financed with tax incentives widens the base of expenditure on which the incentives are calculated; grants reduce the private requirement and therefore the dilution). The second is the correspondence between risk and remuneration: every source should be used in the stretch in which its profile is efficient, because using expensive capital where cheap capital would do destroys value for the promoters, and using rigid capital where patient capital is needed destroys the project.
Section 3
Phased financing and non-monetary capital
The phases. The rational structure for financing uncertainty is capital in conditional tranches: each round covers the stretch up to the next relevant piece of evidence (the de-risking milestones, which in chapters 06 and 07 took the form of TRL levels and gates passed), and the following round is negotiated with that much less risk and that much more value. The logic is the same as the incremental investment of chapter 07, seen from the financier's side: nobody commits today the capital of the whole path, because the information produced along the way changes both the probability and the terms; chapter 10 will show that this structure is formally an option, and will quantify its value. For the project, the discipline of phases has a planning corollary: each tranche has to last until a piece of evidence capable of unlocking the next one, with a margin; capital that runs out halfway between two pieces of evidence is the most dangerous configuration, because it forces you to raise at the point of minimum negotiating strength.
Non-monetary capital. If the function of capital is to reduce risk or add value, then money is only its most liquid form, not the only one. A problem validated by whoever suffers it (chapter 02) spares the project the most failure-prone phase; an already mature technology shortens the TRL path; rare competencies fill the short stave of the barrel (chapter 16); a network of relationships opens channels that money buys slowly and badly; an idea formed over years of sector experience is work already done. Each of these contributions has the economic properties of an investment: it costs whoever provides it, it reduces the need for other people's money, and it shifts the probability of success. Formal recognition of these contributions in the form of a shareholding requires two conditions: a credible valuation (a declared method and a third-party body that applies it, to avoid both the inflation of contributions and their expropriation) and consistency with the corporate structure. Where the two conditions exist, non-monetary capital widens the range of those who can take part in innovation well beyond those who have cash.
Section 4
The repayment order: the honest map of risk
There is one document that tells the truth about the distribution of risk better than any presentation: the order in which the sources recover in adverse cases. When a project stops before the market, what remains to be distributed is the realisable value of what has been built (typically the intellectual property and the documented technology, which can be sold or licensed to third parties: chapter 08 explained why this value is rarely zero, and chapter 12 explains how it is protected). The order of distribution follows the inverse hierarchy of accepted risk: first pure debt, which had an agreed return and no participation in the upside; then the public loan on the terms of the call; then private risk capital; and last the promoter that entered first. Whoever is entitled to the highest return in favourable cases loses first in adverse ones: it is the consistency that makes a financial architecture credible, and verifying it is a due diligence exercise within anyone's reach. A structure in which the promoter declares that it is taking risk but recovers before the investors, or in which a source demands the return of risk without occupying its position, is inconsistent, and the inconsistency can be read in the repayment order before it can be read in behaviour.
A symmetrical note for favourable cases: the same hierarchy explains who participates in the value created. Debt earns what was agreed and nothing more; tax incentives give their return by the tax route, independently of the outcome, plus any shareholding if the structure provides for one; risk capital and the promoter earn through their shareholding, that is, after everyone and only if there is value. Reading the two maps together (who loses first, who earns last) is the quickest way to understand the real incentives of every actor in a financial structure.
Section 5
Case study: mRNA vaccines, or the layering of capital
The development of mRNA vaccines, which culminated in 2020, is a documented case of the layering of sources across decades. The basic research on modified messenger RNA (the work of Karikó and Weissman, later awarded the Nobel Prize in 2023) was financed for years by public and university funds, in a period in which the application was so uncertain that private capital had no reason to be there: it is the Arrow-Bush stretch of section 1, with an enormous social return and an invisible private one. The corporate platforms (Moderna, BioNTech) were built with private risk capital and industrial agreements, on top of that public base: private capital entered when there was a technology to industrialise, that is, at its place in the division of labour. The clinical development and the production scale of 2020 saw the massive entry of grants, advance public orders and procurement instruments (the United States acceleration programme, the European advance purchase contracts), which absorbed the risk of industrialisation at a speed that no private balance sheet would have accepted alone. None of the sources could have replaced the others: the public one would not have built the company, the private one would not have financed twenty years of basic research, and without the emergency instruments the scale would have arrived years later. It is the thesis of this chapter in a single case: the complete path from scientific principle to product in people's arms is financeable only through a layering of capitals, each one in the stretch whose risk is its trade.
In the Volcano method
This chapter is the theory of the six declared sources of the model: Volcano's own capital in the initial stretch that no third party covers ("the part of the path that nobody finances, and the one that decides whether the project exists"), tax capital via the Tax Lease on the TRL 2-5 stretch declared to be the one that no other source covers, public grants and loans as a rule from TRL 5-6 onwards, private capital in a convertible note from TRL 3 onwards, and the pure lender on the positions of minimum risk; the model's entry calendar by TRL is the division of labour of section 2 made graphic. The declared complementarity between private capital and the Tax Lease (the private contribution widens the base of expenditure on which the deduction is calculated) is the example of the first principle of composition. The eight-door theory of capital is section 3 institutionalised: "if it reduces the risk of a venture or adds real value to it, it is capital, whether or not it takes the form of money", with the valuation committee supplying the condition of credibility for the contributions. The repayment order published by the model (pure lender, public loan, private investors, Volcano last, with the tax investor outside the list because its return materialises by the tax route whatever the outcome) is the map of section 4, and its consistency is what gives verifiable content to the formula "we take the first risk": Volcano enters before everyone and recovers after everyone.
Readings
Further reading
- K.J. Arrow, "Economic Welfare and the Allocation of Resources for Invention" (1962): the theoretical foundation of underinvestment in research; a few pages, still unsurpassed.
- M. Mazzucato, The Entrepreneurial State (2013): the documentation of the public role upstream of private technologies; to be read with its criticisms in mind too, so as to form your own judgement.
- P. Gompers and J. Lerner, The Venture Capital Cycle (2nd edition, 2004): the economic analysis of risk capital, including phased financing as a response to informational asymmetries.
- OECD, the periodic reports in the series on R&D tax incentives (the R&D Tax Incentives database and the related country notes): the comparative map of existing tax instruments, useful for placing each national scheme in the family to which it belongs.
Frequently asked questions
Frequently asked questions
If a project is good, should it not find private capital without needing public money?
No, and not through any defect of the project: because of the structure of the problem. Spillovers mean that part of the value created by research cannot be captured by whoever pays for it, and informational asymmetries make the initial stretches expensive even for good projects. Public and tax capital exist to cover the gap between private and social return: their presence in a well structured path is a sign of correct design, not a crutch.
Are non-repayable grants really without cost for the project?
They do not dilute and they are not repaid, but they have costs of their own that have to be weighed: the timescales and the uncertainty of the calls, the eligibility constraints on expenditure, the reporting obligations, and the risk of bending the project to the call instead of the call to the project. The practical rule: grants are integrated into a path that would stand up without them, and they become dangerous when they are its reason for existing.
Why not finance everything with the cheapest source?
Because every source is cheap only in the stretch whose risk corresponds to its profile. Debt is the cheapest source and the most destructive if applied to a research phase: it demands repayments that uncertainty cannot promise. Equity is indispensable in uncertainty and wasted where risk is already low: it pays with a shareholding for what debt would cover at a fixed cost. Financial design is the art of correspondence, not of minimising a single price.
What should you look at to understand who is really taking risk in a structure?
The repayment order in adverse cases, read together with the participation in favourable ones. Whoever loses first and earns last is taking risk; whoever recovers first and has an agreed return is lending; whoever demands the return of risk while occupying the lender's position is describing an inconsistent structure. They are two short lists, verifiable in the documents, and they say more than any declaration.
Disclaimer. This chapter has a training purpose: it describes general categories of financing instruments and their economic logic. It does not constitute tax, legal or investment advice; legislation changes and the applicability of every instrument depends on the specific case, to be verified with qualified professionals.
John F. Kennedy, 1962