Only 8% of European Crowdfunding Is Cross-Border. That's a Market Problem, Not a Rules Problem.

· Policy

Three years into ECSPR, legal passporting works but capital still stays home. The bottleneck is not the rulebook - it is the missing information and decision infrastructure around the offers.

Europe wrote the rulebook for a single crowdfunding market. Three years later, roughly 8% of the volume actually crosses a border.

That single number is the most useful thing to come out of the European crowdfunding policy debate this summer, because it reframes the problem. The question is no longer whether the regulation permits cross-border capital. It does. The question is why capital does not move anyway.

What the three-year review shows

EUROCROWD's status report *ECSPR at Three Years: A Single Market Still in Formation* (7 July 2026, public version 1.1) draws on ESMA data, national reporting and a review of platforms' public disclosures. According to the published summaries, it reports that lending-based crowdfunding accounts for around 58% of European ECSPR volume, equity crowdfunding for only around 12%, and that only around 8% of funding is genuinely cross-border - a share that has barely moved despite platforms being able to passport across the EU.

The report also finds that around 32% of the platforms reviewed show elevated data protection risk, and around 30% show weaknesses in operational resilience, including IT security, continuity and incident handling.

Those figures deserve a stated caveat. EUROCROWD is an industry and policy body, not a supervisory authority, and parts of the platform assessment rest on publicly available information. Absent public documentation is not the same as absent internal controls, and the full methodology and sample should be weighed before the percentages are quoted as settled market statistics.

The directional conclusion, though, is credible and matches what the market looks like from the inside: a licence can make cross-border activity legal, but it does not manufacture investors, deal flow, tax harmonisation, liquidity or trust.

Legal passporting is not commercial passporting

A Danish platform can lawfully offer projects in Sweden, Germany or the Netherlands. It still runs into different tax systems, different company and shareholder-register rules, different investor cultures, translation costs, national marketing practices, thin familiarity with foreign issuers, no local distribution, and genuine uncertainty about how enforcement and investor rights play out in another jurisdiction.

ECSPR harmonised the platform authorisation far more than it harmonised the underlying investment product. That is not a technical footnote. It explains why pan-European platform strategies keep turning out more expensive and less scalable than the business case assumed.

Why EUR 12m is not the answer

Part of the current debate wants the ECSPR ceiling raised from EUR 5m to EUR 12m, partly to align with the higher prospectus threshold under the Listing Act regime. EUROCROWD is sceptical, and the reasoning is hard to argue with: very few ECSPR campaigns come anywhere near the existing ceiling. If the limit is not binding, it is not the bottleneck.

A higher ceiling is genuinely relevant for a handful of scaleups and for mixed institutional/retail rounds. As general growth policy it risks being symbolic. A larger permitted offer size does not create larger investor demand. It does create more complex valuations and ownership structures, and it exposes retail investors to companies that economically resemble small-cap issuances without the analyst coverage or ongoing reporting that comes with them. It also increases the pressure on the KIIS to behave like a mini-prospectus, which it was never scoped to be.

The more defensible position is to accept a higher ceiling in principle and tie it to graduated obligations: stronger requirements on financial history, capital structure, use of proceeds and valuation for larger offers, plus better mandatory reporting after the campaign closes. The central investor protection gap is not only before the investment. It is at least as much the near-silence that follows it.

The platform belongs in the risk chain

If roughly a third of platforms show visible weaknesses in data protection or operational resilience, then analysing only the issuer is an incomplete picture. The platform is part of the investment's risk chain, and most campaign listings implicitly treat it as a neutral pipe.

A separate platform risk layer would look at ECSP licence and home state, passporting status, payment and custody model, whether a bulletin board or other liquidity mechanism exists, wind-down arrangements, history of operational disruption, data protection and public documentation, complaints handling, default or loss data where relevant, and the quality of the platform's own due diligence process.

The part nobody prices in: after the campaign

A Swedish reading of the report - *Equity Crowdfunding: Three Years of ECSPR in Europe* (eAktiebok/Alternativa Nordics, 20 July 2026) - adds a Nordic angle: platform choice matters, cross-border activity is low, a large new shareholder base is an administrative burden, digital share registers help, and platform consolidation is likely.

Read it as an industry observation rather than independent documentation. The publisher plans its own equity crowdfunding service and links the analysis to its share-register products. That does not make the arguments wrong, but the conclusion that the share register is central is also a product message.

The substantive point survives the caveat. A successful campaign does not only produce funding. It produces a governance and shareholder-administration job that lasts for years: nominee structure versus direct ownership, voting rights, information rights, future dilution, drag and tag-along, liquidation preferences, register and investor administration, and how crowd investors get treated in the next round. Plenty of offers look simple at the moment of investment and get complicated later.

Risk disclosure is not decision support

The FCA updated its consumer guidance on crowdfunding on 10 July 2026, highlighting total loss risk, illiquidity, the absence of FSCS cover, and the need to check the platform's authorisation. Sensible, and also a good illustration of the limits of conventional regulatory investor information.

Investors are told the investment is risky, that they can lose their money, that they should research the offer, and that they should understand fees and tax. They are not given an operational method for doing the research. That is the distance between risk disclosure and decision support - between warning someone and equipping them to judge whether a valuation is reasonable, whether the projections hold together, whether founder claims can be documented, whether the terms are skewed, whether there is enough runway, and what has simply been left out.

What is actually missing

The recurring mistake in the European policy debate is treating ECSPR as a regulatory design problem. The three-year evidence points somewhere else: a market and infrastructure problem.

What Europe lacks is not primarily rules. It is comparable investment data, credible post-funding reporting, independent analysis, local distribution capacity, investor education that goes beyond risk warnings, liquidity and exit mechanisms, standards for platform quality, and integration with the rest of the capital market.

Measuring integration by the number of licences issued will keep producing an encouraging chart and a stagnant market. The metrics worth tracking are active platforms, active investors, repeat investment, cross-border volume, issuer survival and realised investor returns.

Europe's problem is not a shortage of crowdfunding offers. It is the absence of a credible, comparable decision infrastructure around them - an independent analysis layer that makes an offer in another country as assessable as one down the road. That layer is what CrowdDiligence is built to be.

Sources

Tags: ECSPR, cross-border, EUROCROWD, platform risk, KIIS