This project report was originally written as a formal deliverable (D4.3) of the DiSSCo Prepare project (Landel et al. 2023b). As such, the document was reviewed by project partners and submitted to the European Commission. Some minor changes have been made to the original deliverable in order to render this version suitable for publishing – this includes the removal of some information on European funding opportunities – however, the authors consider this as the definitive version of the report.

The following text is the formal task description (Task 4.3) from the DiSSCo Prepare project’s Description of work:

Develop the most suitable model for national contributions, based on the existing ESFRI landscape and ERICs; for this task, WP4 will link with WP7 [Governance, Policy & Legal frameworks] regarding legal requirements and DiSSCo Prepare WP8 [Stakeholder engagement & Communication Strategy] for the specialisation graph. This task will especially focus on the national funding mechanisms by assessing the functioning of other existing infrastructures already legally established. Integrating the lessons learned will be essential to determine the contribution model of DiSSCo. In coordination with WP7 and WP8, it will be important to consult the funding mechanisms in the different DiSSCo countries to seek a globally-endorsed model for the implementation and consolidation phases.

The ERIC status created by the European regulation of 25/06/2009 (European Commission 2009) is recognised throughout the Union. Awarded by the European Commission, it allows privileged access to EU-funded calls for projects and benefits from VAT exemption in most member countries. It is based on a commitment from at least three countries (including at least one from the EU), over a period of at least three to five years, supported by a business plan and a provisional budget, which is considered as a commitment from the countries concerned.

ERICs are broad in both mission and service provision and are not governed by any one single applicable funding scheme, although they are highly influenced by negotiations between the EU and Member States. Funding mechanisms should be adapted to each ERIC, although national member contributions are the bedrock of the funding, at least in the early stages.

A benchmarking exercise was carried out over 16 operational ERICs in order to study the main variables of national contribution models. It was a way to identify the main rules to focus on in order to develop a contribution system for DiSSCo. This work was based on the EUR-Lex websites (European Union 2023) containing the official versions of ERICs statutes.

The benchmark includes links to the published statutes of the ERICs; their domain; starting year; number of members and observers; hosting country; GDP (gross domestic product) per inhabitants of hosting country; OECD price level indices of 2019; number of members during preparatory phase; country of the coordinator during preparatory phase; host country contribution (with amount only for France); national cash contributions in 2019; R&D project volumes; income figures (only some identified); other income; host country cash contribution; host country in-kind contribution; other in-kind contributions; and total amount. It also studied information on the membership of the 16 ERICs, such as the type of membership contribution and the rules regulating it, financial figures and data sources.

According to this benchmark, the main rules that exist amongst ERICs statutes are:

A minimum and a maximum fixed annual membership fee;

The ERICs studied were created between 2011 and 2018 and are not all in the same scientific field as DiSSCo (environment) (European Strategy Forum on Research Infrastructures 2021). The minimum total amount of annual contribution identified is €220,000 per year and the maximum €2.2 million per year. These data were collected via Internet searches, including annual reports that are available on the ERIC page of the EC website. Considering that ERIC annual budget reports do not always show clear budgetary divisions in income sources, this information may have some inaccuracies.

Table 1

Although some of the above ERICs are incompatible models for DiSSCo, they reveal some financial rules which could be considered within the DiSSCo member contribution model. These include:

DARIAH – the Digital Research Infrastructure for the Arts and Humanities – imposes an automatic increase of 2% per year of the national annual contribution to compensate for inflation and a monthly calculated fee for new members joining.

EATRIS – the European infrastructure for translational medicine – requests compensation in case of withdrawal of larger countries (> 7% of total contribution).

The BBMRI model could be used for DiSSCo as it provides both fixed and variable contributions. The variable share is based on the GDP of participating countries. An Observer contribution is included (variable share is 30% for the GDP). There is a maximum of 25% for individual countries and international organisations pay a fixed amount that is individually calculated and fixed by the General Assembly (GA).

EPOS statutes contain a detailed formula on how to calculate the membership fee. The contribution is 50% equally fixed and 50% according to GDP contribution. The minimum contribution is 50 k per annum. In principle, this model could be recommended for DiSSCo, with a question mark over the rule of voting rights in proportion to fees, to be negotiated by potential funders.

The EU-OPENSCREEN contribution model combines equal fixed amounts for all member countries (25%) and a variable share of 75% according to GDP per capita (different for member, observer and host countries). The model includes a limit of 50% of total contributions maximum for any individual member. In principle, a model like this could work for DiSSCo.

LifeWatch has implemented a GDP-dependent linear-based contribution model, with a minimum and maximum threshold, although the statutes do not show these thresholds in detail. This constitutes the simplest model for calculating mandatory annual contributions and, therefore, it is a possible model for DiSSCo. The LifeWatch budget is always fixed for 5 years.

Takeaways – Contribution models are typically intended to meet the following objectives:

The different countries’ shares should be calculated according to a transparent methodology relevant to the purpose of the ERIC, such as their population size, economic power, number of potential users, R&D spending or shall be equal for all;

As the above variety demonstrates, there is no classic model which ERICs can copy and paste or which all countries have agreed on. Therefore, each ERIC needs to develop its own specific model, adjusted to its specificities and needs. Indicators relevant to the member contribution calculation can include:

Population size is a good first indication in approaching comparability as it is easy to identify for most countries. It does not, however, reflect the economic power of a country nor its R&D capacity.

Economic power is typically represented by Gross Domestic Product (GDP) or Gross National Product (GNP). Both represent the total market value of all goods and services produced over a certain period. However, they are calculated in slightly different ways. GDP is the value of the finished domestic goods and services produced within a nation's borders. On the other hand, GNP is the value of all finished goods and services owned by a country's citizens, whether those goods are produced in that country or not. While GDP limits its interpretation of the economy to the geographical borders of the country, GNP extends it to include the net overseas economic activities performed by its nationals. GDP – sometimes called GNI (Gross National Income) – is the more frequently used of the two indicators. As countries might expect annual differences in GDP, some ERICs prefer to calculate the average GDP over a number of years (typically 3 years).

Research and Development: R&D indicators are divided into gross:

gross domestic expenditure on R&D (GERD);

All these figures are available for EU and OECD countries. GERD includes expenditure on research and development by business enterprises, higher education institutions, as well as government and private non-profit organisations. In order to make the figures more comparable, GERD is often expressed relative to GDP or in relation to population. The ratio of GERD to GDP is also known as R&D intensity. As most R&D expenditure is covered by industry and different countries have a different industrial base, countries might not be comparable. Alternatively, one might look at the higher education sector or private non-profit sector. Finally, it is possible to use R&D expenditure by source of funds, which describes the origin of the R&D funding for a statistical unit. Performer-based reporting of the sums which one unit, organisation or sector has received from another unit, organisation or sector for the performance of intramural R&D. R&D funds are identified with two criteria: there must be a direct transfer of resources and this transfer must be both intended and used for the performance of R&D. Source-based reporting of extramural expenditure which are the amounts a unit, an organisation or a sector reports having paid to another unit, organisation or sector for the performance of R&D.

Number of users per country: very often institutions would like to use the number of potential users of a country or, if that is not available, the number of researchers. R&D personnel consists of all individuals employed directly in the field of R&D, including persons providing direct services, such as managers, administrators and clerical staff. R&D researchers can be employed in the public or the private sector - including academia - to create new knowledge, products, processes and methods, as well as to manage the projects concerned. Countries with a stronger industrial base, therefore, might have a higher number of R&D staff, making it more difficult to estimate potential users of a RI, who currently come mainly from academia or the public sector.

For most countries, reliable past and current statistics are available from EUROSTAT, OECD.Stat and World Bank Open Data. Additional variables, not linked to relative financial statistics, include:

Introduce a threshold: as some EU countries can be very small (e.g. Luxembourg) or very large (e.g. Germany), consortia developing a contribution model want to implement thresholds to ensure fairness. This might make sense when the formula developed is only based on population size or economic power alone. In this case, it is possible to introduce absolute or relative thresholds. Examples are minimum and/or maximum amounts or percentages. Another possibility is to split the contribution into fixed and variable amounts. In the latter case, there might still be a difference in size (like BBMRI or EATRIS) or a fixed amount (like EU-OPENSCREEN). To reduce the risk of larger countries having to pay the lion’s share of the total member contributions, some ERICs have introduced a mechanism of maximum share (BBMRI and EU-OPENSCREEN), with a redistribution of the surplus across other participants.

Host country premium: The host country premium or the additional contribution of the state of incorporation (Statutory seat) can be a delicate matter. Comparisons show a very diverse picture with no common rule. The minimal amount should be able to cover the expenses for the central office (without personnel), incorporation, insurance, taxes etc. In rare cases, several countries pay a premium for hosting specific common services (like ICOS (the Integrated Carbon Observation System) – and BBMRI (Biobanking and Biomolecular Resources Research Infrastructure)); this premium should be calculated transparently. Ultimately, it will depend on negotiation with the potential host country/countries.

Inflation rate: as inflation rates can be high, it is recommended to include a consideration for inflation in the calculation. Inflation is the increase in the general level of prices of goods and services in an economy; the reverse situation is deflation, when prices decrease across the board. Inflation and deflation are usually measured by consumer price indices or retail price indices. Within the EU, a specific consumer price index has been developed: the harmonised index of consumer prices (HICP). One possibility is either a fixed or a flexible percentage, calculated, based on the previous year’s figures. It would, therefore, be a case of choosing the flexible percentage or using the index relevant for the host country (as most goods and services would be covered there) or an average of participating states (which will need to be recalculated with each new member) or the EU average. However, accounting for inflation each year may be not acceptable by member states if their commitment to ERIC’s funding budget is on a 5-term year.

Withdrawal of member countries: withdrawal of member countries will always represent a risk for ERICs as the money might have already been allocated or personnel hired. Therefore, reasonable measures need to be taken to notify in advance (1-2 years ahead of withdrawal for less equipment-based RIs like DiSSCo) and/or foreseen fees (e.g. in percentage of the annual contribution) for premature departure (25-30% per annum).

Embedding international organisations

ERICs allow international organisations (IO), for example the United Nations (UN), to be members or observers. In the case of the UN, it requires the permission of the 193 members. For embedding IO as observers, the Director General (DG) can decide and there is no rule as to how to calculate their contribution. It can be directly negotiated between the IO and the ERIC’s GA. It is not mandatory that it is specified in the statutes. For instance, it can be decided during the first GA.

Minimum number of members to guarantee funding the RI

If some members leave the RI over the course of its implementation, it is possible to raise the topic during the GA to decide if national contributions should be increased or the level of service provision decreased. Therefore, the ERIC cannot be launched with fewer than five countries.

In case of withdrawal of a member during the first 5 years of operation, the country concerned will have to pay for all 5 years of their initial commitment. This rule is guaranteed by the ERIC regulation. If a member does not want to pay, the ERIC would have the right to go to court. In case of withdrawal of a member after this 5-year period, the GA should be notified in advance: the notice period should be included in the Statutes.

Minimum contributions expected from observers

There is no rule for the minimum contribution expected from observers, although the details of observers’ subscription to the ERIC should be detailed in the ERIC’s statutes. It is realistic to ask 1/4 to 1/3 of the full membership fee. If observers have the same rights to access the services as members, 1/3 of the regular member contribution amount may be appropriate.

Definition of in-kind: statutes or service level agreements (SLAs)

This depends on the expectations from the ERIC towards its member-institutions. If the in-kind contribution is strategically important, it is possible to write in the statutes that there will be an expectation on the participating institutions to contribute in-kind. It is also possible to add in the introduction of the ERIC statutes that institutions will be asked to contribute in-kind.

Voting of the budget

Budgetary cycles should not be specifically mentioned in the statutes. Each year the budget is drafted and, as long as the GA agrees with the proposal, it will be actioned. In the case of higher investments (around 1/3 of the regular member contribution can be a threshold), it is possible to prepare a 2-3-year budget.

States joining mid-year

A rule should be included in the statutes in order to clarify how it would work in case a country joins the ERIC outside of the budgetary cycle. In most cases, countries commit to funding the ERIC for 5 years after joining.

Joining is effective as of the date after the GA has agreed to it, in writing. It is likely that the decision to welcome a new country will be taken during the GA.

Financial penalty in case of early withdrawal

A rule is in place in case a member country withdraws from the ERIC before the end of the first 5 years. A good option is to negotiate early notification in case of a decision to exit the ERIC.

Non-EU countries

Non-EU countries can be assigned the same membership fee calculation. Sometimes, EUROSTAT only holds data for EU countries. In that case, it is possible to use equivalent data from the World Bank or OECD.

ERICs’ eligibility to access loans

ERICs can access loans like any private company. In practice, this requires the ERIC to demonstrate its bankability for a lender. The GA is responsible for approving the loan request. As it currently stands, the only case of an ERIC requesting a loan concerns the European Spallation Source ERIC (ESS). The whole construction cost amounted to €2 billion and required a loan guaranteed by the European Investment Bank.

In the case of an ERIC like DiSSCo, the decision should be taken 10 years before the request for the loan.

In case of reserves/cashflow

Reserve authorisation will depend on the financial and monetary rules of the country with statutory seats. The rules will vary from one country to another. In some countries, ERICs are seen as a private organisation and should, therefore, follow private financial rules. In other countries, they are seen as public institutions. Typically, in the EU, public organisations are advised against amassing cash reserves.

DiSSCo’s initial developments implemented through EU-funded projects (ICEDIG ("Innovation and Consolidation for Large Scale Digitisation of Natural Heritage", EU Horizon 2020 grant agreement No. 777483), MOBILISE COST (EU COST Action CA17106 on “Mobilising Data, Experts and Policies in Scientific Collections”), DiSSCo Prepare and SYNTHESYS+ (“Synthesis of systematic resources”, EU Horizon 2020 grant agreement No. 823827) represent an estimated budget of more than €13 million. This budget mainly includes EU funding as well as in-kind contributions from participating institutions. In February 2023, DiSSCo will stop benefitting from these European funding programmes and enter into its transition phase (see Fig. 1). During the transition period, DiSSCo ERIC’s statutes and national annual contributions will be discussed amongst its future Members and Observers. The national contributions will represent the fixed annual budget of the RI. It is the funding on which much of the expenditure presented within this document will rely. The national contributions described within this document are intended to cover core DiSSCo activity for five years: the fixed costs of managing the DiSSCo Central Hub (€1.4 million per year) (Landel et al. 2023a) and provision of core services. This assumption is based on an incremental and realistic approach. DiSSCo will first deploy a basic team with tools to improve access to NSC by helping institutions to follow European shared innovations.

Fig. 1

Four principles underpin the DiSSCo annual membership calculation methodology:

National membership fees shall be fair and equitable. They shall reflect the resources devoted to science and technology amongst DiSSCo member countries’ population size and annual spending on R&D;

These principles are supported by the five following recommended rules:

Host Country annual membership will represent 25% of the annual cost of the Central Hub Office;

The calculation of national annual contributions to research infrastructures should strive to fairly distribute financial commitment amongst its members.

Two possible options on which to base the DiSSCo ERIC formula to calculate the membership fee have been defined: both consider the GDP, R&D spending and population size of potential DiSSCo Member States. Countries do not rank identically if gross or per capita values are considered. If two gross values are used – for example GDP and GERD – the correlation coefficient is high and, therefore, the variables are tightly clustered, meaning a less equal distribution of funding. Furthermore, countries rank differently with respect to GDP and GERD. As a consequence, mixing gross and per capita in the formula lessens the difference.

Some countries could have annual differences in GDP. To mitigate against this, DiSSCo ERIC national contributions can be calculated on the average GDP and GERD over 3 years (see Appendix 1).

The indicators preselected, GERD/cap, GDP, GERD, GDP/cap, do not vary widely over time. Whilst they did grow between 2017 and 2019, no drastic variation was identified. This means that funding connected to these variables should be stable.

The size of the Natural Science Collections (NSCs) of the participating countries was not selected as an indicator. The figures on the Collections are provided by institutions, potential members and service providers of the ERIC. Meanwhile, the contribution model requires indicators that come from external stakeholders. Population size, GDP, R&D expenditure (etc.) figures are provided by EUROSTAT, World Bank and OECD, therefore, avoid any potential conflict of interest. For some institutions, the figures on the size of collections still need to be clarified, so this is not a reliable indicator.

It could be argued that the most accurate parameter for establishing member contributions would in fact be the size of the scientific community impacted by the establishment of DiSSCo RI. The problem being, however, that we do not, at this stage, know how to calculate this figure. It would require participating governments to carry out a detailed and exhaustive study of the outreach of their Natural Science Collections which would be an extremely costly and timely exercise. Furthermore, it would require each participating government to carry out the same exercise, which would be a huge challenge given the different priorities and budgets of each government.

As the DiSSCo national contribution model is flexible, it was possible to test the different options to experiment with the range of values obtained, as shown in the following three combinations:

GDP and GERD;

This section aims to model the effects of these different options on the DiSSCo annual membership fees. It imagines a scenario whereby all 11 DiSSCo Funders Forum members (the Funders Forum being ministerial representatives of potential DiSSCo national funders during its preparation phase) sign the DiSSCo ERIC statutes and annually contribute to funding the Research Infrastructure. Using this scenario, the model simulates the total annual budget of the RI. Finally, it will simulate the medium-term effects of inflation on these contributions.

One possibility for the DiSSCo annual contribution model is to add inflation as a factor of evolution of the membership fees. This may not be indexed on an annual basis, but instead could evolve every five years (minimum length of commitment).

Inflation is highly unpredictable: as the working hypothesis to test the effect of inflation, we assume 2% annual inflation rate and DiSSCo's statutory seat being in the Netherlands. If the costs of the RI rise due to inflation, it will likely be connected to the inflation rate of the host country. On that basis, it is possible to model the impact of inflation on DiSSCo’s income, as shown in Fig. 6.

In line with inflation, three main figures would evolve accordingly:

The baseline fee (€50,000);

If the RI is launched in 2024, the first indexation of its fees would be in 2029. According to the data connected to the graph above, the following evolution may happen:

Indexed baseline fee: €55,200;

If we take the median of annual contributions in 2024 (starting year) and the median of annual contributions in 2029 (+5 years), we obtain the graphs below (Fig. 7, Fig. 8 and Fig. 9). We see that inflation does not have a major impact on the evolution of each country’s annual membership fee. On average, the contributions would be increased by 10% between 2024 and 2029.

Fig. 7

Fig. 8

Fig. 9

As shown above, the most balanced options are:

Option B: 40% GDP and 60% GERD/cap;

We recommend these two options to share with DiSSCo potential future funders, summarised in Table 8. As a lot of data can be extracted from this flexible formula, a selection has to be made on the basis of predefined criteria.

The graph below illustrates Table 8. The annual contributions remain stable across Options B and C. Except for a few countries (with higher GDP/cap), choosing between the two options will not greatly impact the budgets of the DiSSCo future members.

Fig. 10