Innovation needs everyone.

ResearchBehavioural foundations

Whether people take part is not decided by money alone. It is decided by knowledge, trust, habit and by how the decision is put to them.

01Why households are absent

It is not a lack of money that holds households back. What is missing is knowledge, access and an occasion to begin at all.

Those who take part in capital markets differ less in income than in knowledge: financial literacy is closely linked to stock market participation, even once other characteristics are accounted for.4 In Germany there are further explanations that do not lie in wealth but in habit, information costs and the sense that securities are meant for other people.9

Every station between first interest and actual investment costs attention, time and self-confidence. Where one of these stations is too demanding, participation ends, regardless of whether someone wanted to invest. That is precisely why the design of the procedure is not decoration but the real lever.

Where participation endsFrom everyone who could in principle take part to actual participation: five stages, schematic
Stations between interest and investment Ten people on the left stand for everyone who could in principle take part. They pass through five stations: attention, understanding, trust, decision, settlement. At each station some leave, drawn in grey and labelled never heard of it, too complicated, no trust, postponed, abandoned. One person remains at the end. Below each station is what the procedure sets against this.
02Trust

Nobody can fully assess an early-stage venture. Trust is the mechanism that allows a decision nonetheless.

Trust reduces complexity: it allows people to act although far from everything can be known.2 Empirically this is no side issue. People with less trust in others and in institutions take part in capital markets less often and invest smaller amounts, irrespective of wealth.3

For an early-stage venture the gap between the knowledge required and the knowledge available is particularly wide: there is no track record, no price, no benchmark. That gap cannot be closed with more text. It can only be bridged by making the procedure itself traceable: standardised profiles, an expert review without any financial interest of its own, and a named supervisory authority.

Key point

The test for CIVINCO is therefore not whether people know enough, but whether they can trust the procedure within which they decide.1

The knowledge gapWhat an assessment requires, what a person can manage
Knowledge gap and procedure Three rows, all starting at the same point. At the top, in grey, seven fields for the knowledge that assessing a venture yourself would require: technology, market, team, competition, law, valuation, time. Below, in black, only two fields for what a person realistically brings: an understanding of the product and an impression of the team. A vertical line marks that limit. The remaining gap is filled by the third, blue row: standardised profile, admission by the EAC, documented steps, supervision.
03Social norms

What many people do becomes the benchmark. That helps participation and can damage it at the same time.

Feedback about the behaviour of others measurably changes one’s own behaviour. But it works in both directions: those below the average move towards it, those above it fall back. Only a sign of approval prevents that rebound.7 In large field experiments such feedback reduces consumption reliably, but by a modest magnitude.8

For CIVINCO this implies a precise distinction. It may be visible that people take part, because that is what creates the ordinariness on which participation rests. Amounts, rankings and comparisons between individuals must not be visible, because that is what creates competition over stakes. The difference between a norm and a game lies exactly here.1

How a norm emergesWhat is visible is that people take part, not how much they put in
Spread of participation in three stages Three dot fields of equal size side by side, 48 dots each for 48 people. From left to right more and more dots are filled blue: first 4, then 12, then 24. Blue means that a person takes part; how much they invest is not shown, all dots are the same size. Schematic representation.

The distinction is written into the model: your own position is visible, other people’s amounts are not. See No game mechanics.

A row of houses with plants on almost every balcony; on a balcony that was still empty, someone sets down the first pot while the neighbour hands over a watering can.
Social norms

Participation you can see. What is visible is that people take part. Not how much.

04From intention to action

Between agreement and action lies friction. Reducing it is permitted; taking the decision away is not.

The effect of defaults is well documented in retirement saving: once participation becomes the default setting, it rises sharply, and many people subsequently stay with the preset values.5 A milder variant is documented as well: where employees are required to make an active decision, without any direction being suggested, participation rises markedly compared with a pure opt-in procedure.6

For a risk investment the first variant is ruled out. A participation people would have to opt out of would be neither defensible in supervisory terms nor compatible with the voluntariness that CIVINCO makes a condition. What remains is the second: a standardised path, a low hurdle, a conscious decision.

In practice: one uniform process for all ventures, comprehensible profiles, small amounts, clear deadlines, and at the end a confirmation a person has to give actively. Whatever makes the decision easier belongs here. Whatever replaces it does not.

Three ways to pose a decisionWhat happens when someone does nothing

Opt-in

Those who do nothing do not take part. The whole hurdle lies with the individual.

high hurdle

Default

Those who do nothing take part. Anyone who does not want to must opt out.

ruled out for a risk investment

Active choice

No one is enrolled by default. The path is the same for everyone; each person gives the confirmation themselves.

the path in CIVINCO

The findings on defaults and active choice come from retirement saving.5 6 Direct evidence for risk investments is lacking.

05Incentives beyond behaviour

Behaviour explains the hurdle. It does not replace the reason why someone puts money into a risky venture.

The previous sections show how a procedure lowers friction and supports trust. They do not explain why someone puts part of their money into a venture that is likely to fail. Skipping that question means building the model on idealism.

Three levers are available. Limiting risk is built into the model. Co-decision is present and does not carry on its own. The third lever, the tax treatment of return and loss, lies outside the model: it is a decision for the legislator.

Three leversWhat lies within the model and what does not
built into the model

Limit risk

caps, small amounts, diversification

present, not enough on its own

Co-decision

choosing for yourself what to invest in

a decision for the legislator

Return and loss

tax treatment of the micro-investment

The federal government already supports private early-stage investment. INVEST pays an acquisition grant of 15 per cent of the issue price and, at an exit, 25 per cent of the capital gain.10 The Income Tax Act makes such grants tax-free up to 25 per cent of the acquisition cost.11 The statutory framework is therefore wider than the directive makes use of. As a template INVEST nevertheless does not fit CIVINCO: it requires a minimum investment of 10,000 euros per investor and thereby excludes micro-investments.

At amounts between 50 and 500 euros the usual tax instruments barely work. A tax-free capital gain would come to nothing, because the saver’s allowance of 1,000 euros as a rule already covers it. A loss deduction only helps those who have other capital income to set it against.11 What works is therefore only what applies at the moment of investment: a grant or a refundable credit. That is not a detail but the economic design decision of the model.

Two European programmes show that this works without a minimum threshold. The British Seed Enterprise Investment Scheme grants 50 per cent income tax relief on up to 200,000 pounds a year and names no minimum amount.12 France grades the rates by degree of innovation: 50 per cent for young companies with breakthrough innovation, 40 per cent for those with an impact objective, 30 per cent under the general regime, capped at 50,000 euros of subscriptions per person and year.13

The European state aid framework already accommodates this. Since 2023 the General Block Exemption Regulation exempts tax incentives for private investors who are natural persons from the notification requirement, with up to 50 per cent of the investment, up to 65 per cent in assisted areas and up to 30 per cent where the finance is provided through a financial intermediary.14 The catch leads back to the question of how the participation is held: on the direct route only newly issued ordinary shares with full risk are eligible, held for at least three years. A subordinated loan would fall outside. And the regulation in force is limited to the end of 2026. See How the participation is held.

Incentives comparedRelief at the time of investment, as a percentage of the amount invested
  • INVESTGermany, grant. Minimum investment 10,000 euros10
    15%
  • France, general regimeup to 50,000 euros of subscriptions per person and year13
    30%
  • France, impact objective
    40%
  • France, breakthrough innovation
    50%
  • United Kingdom, SEISup to 200,000 pounds a year, no minimum amount12
    50%
  • EU state aid framework, ceilingup to 65 per cent in assisted areas, up to 30 per cent through financial intermediaries14
    50%

INVEST is a grant; the other rates are tax reliefs. The comparison is between rates, not their effect in an individual case.

What this section does not do is choose an instrument. That is a political decision with budgetary effect and belongs in the phase 0 review. Two limits apply regardless: an incentive must not lead anyone to risk more than they can bear, which is why the caps remain the first rule and not the second. And deadweight effects are to be measured, not asserted. See Evaluation and implementation.

06From finding to rule

Every finding has its counterpart in the procedure. Otherwise it would merely be a citation.

From finding to ruleWhat the research suggests and where it appears in the model
FindingDesign ruleIn the model
Knowledge is unequally distributed and costly to acquireStandardised, reviewed profiles instead of your own researchEAC, section 07
Trust substitutes for missing knowledgePublicly accountable procedure, documented stepsArchitecture, section 02
Small steps lower the hurdleMicro-investments with fixed capsThree pillars, section 06
Social comparison can backfireParticipation is visible, the amount is notNo game mechanics, section 12
Defaults would not be permissible hereActive choice: low hurdle, conscious decisionProcess, section 03
The crowd prefers the familiarIntermediation and curation according to published rulesThe condition, section 13

The findings come mostly from adjacent fields: stock market participation, retirement saving, energy consumption. For participation in early-stage innovation finance the direct evidence is missing. That is precisely why a pilot with evaluation is part of the proposal and not an appendix. See Evaluation and implementation.

07Sources

Sources for this page. Numbered as in the text.

The superscript numbers in the text refer to this list. The derivation of the model is set out in the working paper.

The model in detail

Steiner, T. M. (2026). Collective Risk-Sharing in Innovation Finance: An Institutional Model of Citizen Investorship. Working paper, SSRN. Under review.DOI 10.2139/ssrn.6867205

  1. Steiner, T. M. (2026): Collective Risk-Sharing in Innovation Finance: An Institutional Model of Citizen Investorship. SSRN working paper. doi.org
  2. Luhmann, N. (1968): Vertrauen. Ein Mechanismus der Reduktion sozialer Komplexität. Stuttgart: Enke.
  3. Guiso, L., Sapienza, P. & Zingales, L. (2008): Trusting the Stock Market. The Journal of Finance, 63(6), 2557–2600.
  4. van Rooij, M., Lusardi, A. & Alessie, R. (2011): Financial Literacy and Stock Market Participation. Journal of Financial Economics, 101(2), 449–472.
  5. Madrian, B. C. & Shea, D. F. (2001): The Power of Suggestion: Inertia in 401(k) Participation and Savings Behavior. The Quarterly Journal of Economics, 116(4), 1149–1187.
  6. Carroll, G. D., Choi, J. J., Laibson, D., Madrian, B. C. & Metrick, A. (2009): Optimal Defaults and Active Decisions. The Quarterly Journal of Economics, 124(4), 1639–1674.
  7. Schultz, P. W., Nolan, J. M., Cialdini, R. B., Goldstein, N. J. & Griskevicius, V. (2007): The Constructive, Destructive, and Reconstructive Power of Social Norms. Psychological Science, 18(5), 429–434.
  8. Allcott, H. (2011): Social Norms and Energy Conservation. Journal of Public Economics, 95(9–10), 1082–1095.
  9. Ebert, S., Grote, M. & Laudenbach, C. (2019): Zum Rätsel der Aktienmarktteilnahme in Deutschland. Frankfurt School of Finance & Management and Goethe University Frankfurt, commissioned by Deutsche Börse AG. uni-heidelberg.de
  10. INVEST — grant for venture capital, funding directive of 06.03.2024 (BAnz AT 20.03.2024 B2), items 4.1.3, 5.1 and 5.2. bafa.de
  11. German Income Tax Act, § 3 no. 71 (tax exemption of the grant) and § 20(6) and (9) (loss offsetting, saver’s allowance). gesetze-im-internet.de
  12. GOV.UK: Venture capital schemes, tax relief for investors (Seed Enterprise Investment Scheme), as at 06.04.2026. gov.uk
  13. Bulletin officiel des finances publiques: Réduction d’impôt au titre des souscriptions au capital de JEI, JEIR et JEII, BOI-IR-RICI-110-20, as at 27.08.2026. bofip.impots.gouv.fr
  14. Regulation (EU) No 651/2014 (General Block Exemption Regulation) as amended by Regulation (EU) 2023/1315, Article 21a and Article 59. eur-lex.europa.eu