H2020 Monitoring Report: Keeping Our Eyes on the Horizon

Keeping our eyes on the HORIZON MONITORING FLASH SERIES A HORIZON 2020 MONITORING REPORT #HorizonEU

Research and Innovation Keeping our eyes on the Horizon – Monitoring flash series European Commission Directorate-General for Research and Innovation Directorate A — Policy & Programming Centre Unit A.2 — Programme Analysis & Regulatory Reform

Contact Martina Kadunc – Team Leader Impact Monitoring Nelly Bruno – Team Leader Programme Analysis and Evaluation Email [email protected] [email protected] [email protected] [email protected]

European Commission B-1049 Brussels

Manuscript completed in 2020. First edition.

The European Commission is not liable for any consequence stemming from the reuse of this publication. The views expressed in this publication are the sole responsibility of the author and do not necessarily reflect the views of the European Commission. More information on the European Union is available on the internet (http://europa.eu). PDF ISBN 978-92-76-11331-7 doi:10.2777/303024 KI-02-19-806-EN-N Luxembourg: Publications Office of the European Union, 2020

The reuse policy of European Commission documents is implemented based on Commission Decision 2011/833/EU of 12 December 2011 on the reuse ofof CommissionCommission documentsdocuments (OJ(OJ L L 330, 330, 14.12.2011, 14.12.2011, p. p. 39). 39). Except Except as otherwise otherwise noted, noted, the the reuse reuse of of this this document document isis authorisedauthorised underunder a Creative Commons Attribution 4.0 International (CC-BY 4.0) licence (https://creativecommons.org/licenses/by/4.0/). This means that reuse is allowed provided appropriate credit is given and any changes are indicated. For any use or reproduction of elements that are not owned by the European Union, permission may need to be sought directly from the respective rightholders. Image credits: cover: © toonsteb # 279375161, 2020. Source: stock.adobe.com EUROPEAN COMMISSION

Keeping our eyes on the Horizon MONITORING FLASH SERIES A HORIZON 2020 MONITORING REPORT

edited by Martina Kadunc and Nelly Bruno

Directorate-General for Research and Innovation 2020 Programme Analysis & Regulatory Reform

Investments under the EU’s R&I framework programme are crucial for Europe’s future

Research and innovation (R&I) has demonstrated its importance in the context of the coronavirus pandemic and will continue to play a central role in Europe’s recovery, preparedness and resilience. New knowledge and solutions developed through the EU’s R&I framework programme will help us to make Europe the first climate-neutral continent in the world by 2050 and to grasp the opportunities of the digital age. The new programme, Horizon Europe, is integral to an ambitious recovery plan and a modernised 7-year budget for the EU.

These five ‘monitoring flashes’ (series title: “Keeping our eyes on the Horizon”) present detailed analysis of key aspects of the programme’s development in recent years, such as participation by country and its contribution to sustainability issues. They show that Horizon 2020 is already helping to build a more sustainable future: four in every five projects tie in with specific sustainable development goals. Programme activities are drawing more and more universities and companies into a massive network of over 1.5 million collaborations worldwide. The programme is delivering quality results in the form of patented inventions with above-average market value and scientific publications that are quoted more than the world average.

The example of the ongoing pandemic remains a stark reminder that R&I is at the forefront of efforts to address the most pressing global challenges. However, even with the best intentions and appropriately large-scale investment, breakthroughs are rare and take time. The availability of capacities to respond quickly and effectively today depends on visionary investments made yesterday. More than ever, the EU needs to keep a close eye on the investments it is making now. Even more importantly, we must monitor the results and impacts for science, the economy and society at large.

The monitoring flashes are part of an effort to modernise the monitoring and evaluation of the R&I framework programmes. They complement real-time data on programme implementation and results, as provided by the Horizon dashboard and the Horizon results platform, and lay the ground for the Horizon 2020 ex post evaluation in 2023 and the Horizon Europe interim evaluation in 2024. The new approach will also involve a novel monitoring framework for Horizon Europe, based on ‘key impact pathways’, whereby streamlined indicators will help us track the programme’s effectiveness in delivering scientific, economic and societal progress.

Programme monitoring is central to the European Commission’s ‘better regulation’ agenda, which puts evidence-based analysis at the heart of EU policy action (design, implementation and redesign). Timely strategic analysis of EU programmes in their wider regulatory and socio-economic contexts makes it possible to identify, at an early stage, driving forces and opportunities for future action (including on regulation, through ‘innovation deals’ to explore regulatory burdens and implement the innovation principle in the EU).

2 Horizon 2020 is an attractive and efficient R&I programme, which was quick to respond to the COVID-19 pandemic

Horizon 2020 was quick to respond to COVID-19. The first European case was reported in France on 24 January 2020. Just 7 days later, Horizon 2020 emergency research funding of €100 million was granted for research into the coronavirus. Since then, the EU has promised to invest around €1 billion by the end of 2020 to tackle COVID-19 and its consequences. The EU R&I community, coordinated by the Commission, is providing an ongoing, fast and robust response to the crisis.

Horizon 2020 remains an attractive programme. Since its launch in 2014, nearly 250 000 eligible proposals have been received in response to over 700 calls for proposals — this is twice as many per year than under the predecessor programme (FP7)1. However, the high level of interest has to be seen against the background of a decade-long stagnation in public expenditure on research and development (R&D) across EU Member States2, whereas business expenditure has risen steadily3. Also, EU resources are limited and Horizon 2020 has been able to provide funding for only one in eight proposals (11.9%) submitted under the programme; this success rate is well below what was achieved under FP7 (18.4%). Unfortunately, of every four of the proposals evaluated by 36 000 independent experts as being ‘of high quality’, three could not be funded.

Nevertheless, when demand exceeds resources, Horizon 2020 helps to open the door for other opportunities. Nearly 25 000 individual entities that could not receive funding due to budget limitations have been awarded a ’seal of excellence’, a quality label that helps them access alternative financing, e.g. from the European Structural and Investment Funds or from private sources. Some 139 entities have received a new COVID-19 seal of excellence.

To date, Horizon 2020 has involved over 1.5 million individuals4 in over 30 000 projects benefiting from investment of €56 billion, of which 30% has been for climate-related and 4.9% for biodiversity-related actions.

€56.4 bn 30 464 30% 36 039 22% of EU R&I investment grants signed with an of investment related to entities, including of investment going to in Horizon 2020 projects average of 5 partners climate action 66% newcomers and SMEs (target: 20% to date per project 15 580 SMEs under ‘societal challenges’ and LEITs5)

€1 bn €440.6 m 103 547 3 090 of EU R&I investment of EU R&I investment grants signed to fight additional projects entities funded through by the end of 2020 to since January 2020 to COVID-19 reoriented to fight projects to fight tackle COVID-19 tackle COVID-19 COVID-19 COVID-19

1 45% of the proposals (accounting for 34% of the budget requested) were for projects under Marie Skłodowska-Curie actions and with the European Research Council, followed by the SME instrument/European Innovation Council pilot and the ICT and health priorities. 2 Eurostat data until 2019 (before the COVID-19 crisis). 3 For more information, see 2020 report on scientific, research and innovation performance of the EU. 4 At least 800 000 researchers are reported as being involved (36% of whom are women), and 700 000 non-researchers (47% women). 5 ‘Leadership in enabling and industrial technology’ priorities.

3 A large share of the investment (€21 billion) is for projects under the ‘excellent science’ pillar, mainly those run by the European Research Council (ERC; €11 billion) and under the Marie Skłodowska-Curie actions (MSCAs; €5 billion). Investment in the ‘societal challenges’ (SC) pillar is also €21 billion, much of it (€5 billion in each case) for SC1 (health, demographic change and wellbeing) and SC4 (smart, green and integrated transport). A total of €13 billion has been allocated under the ‘industrial leadership’ pillar, mainly for information and communication technologies (€6 billion) and nanotechnologies, manufacturing, advanced materials and biotechnologies (€3 billion).

SEWP: Spreading excellence and widening participation SWAFS: Science with and for society

The main beneficiaries of Horizon 2020 funds are secondary and higher education institutions (39% of budget allocated to 2 461 entities), followed by private for-profit companies (28%, 23 839 entities) and research organisations (26%, 3 102 entities). Overall, 15 580 SMEs are involved in the programme. Almost 66% of Horizon 2020 participants are newcomers6, with private for-profit companies making up a large majority.

As a result of simplification efforts under Horizon 2020, grant agreements are now signed 1.7 times faster, taking an average of 184 days as against 313 under FP7. Overall, 89% of grant agreements are signed within the 245-day ‘time to grant’ target7.

6 i.e. successful applicants who did not apply under FP7. 7 The ‘time to grant’ is the period between the closing date of the call and the signing of the grant agreement (the official start of the project). Under Horizon 2020, the Commission has committed to signing agreements within 245 days (8 months) for all calls except ERC calls.

4 Horizon 2020 strengthens R&I capacities through cooperation across the EU and beyond

Horizon 2020 has a broad international outreach, involving entities in 166 countries. 90.1% of the funding has gone to organisations or individuals in the EU8 and 8.6% to ‘associated countries’9.

Monitoring flash 1 (country participation) gives a detailed analysis of applications and participation from entities in EU Member States. It shows that half the applications involve entities in only five countries: Germany, the United Kingdom, France, Spain and Italy. These countries also have big national R&I systems, hosting 60% of all EU scientists and engineers attracting over 70% of all R&D investment in the EU.

Under Horizon 2020, the proportion of funding going to ‘EU-13’10 countries has grown slowly and is now in line with their share of EU-wide R&D investment (around 4%). The number of applications from these countries is rising slightly, but it remains relatively low given their scientific potential (i.e. the number of scientists and engineers in the population). Nevertheless, when it comes to the ratio between the number of applications and the number of scientists and engineers, four of the countries (Cyprus, Slovenia, Malta and Estonia) are among the top 5 in the EU, together with Greece. Still, individual applicants from EU-13 countries apply 1.7 times less frequently than their EU-15 counterparts.

Overall, the analysis indicates that the more a country invests nationally in its R&D capacity, the more funding it receives from the programme. This tallies with the findings of the 2020 science, research and innovation report11, which point to a positive correlation in the EU between the level of national R&D investment and scientific quality (a core evaluation criterion for framework programme funding).

The level of participation in Horizon 2020 is a complex issue. Differences in R&I performance among Member States are determined by a whole range of factors beyond the control of the programme, including national priorities, the level of private and public investment, the availability and quality of infrastructure, human capital and skills, access to finance, the support measures in place, the regulatory framework, etc. These require national policy mixes that are tailored to specific challenges. Although several Member States are trying to improve their capacities (e.g. through the European Structural and Investment Funds), further efforts and synergies between the EU, national and regional levels are needed to ensure well-functioning, efficient and impactful national R&I systems within the European Research Area (ERA).

A key aspect of ‘EU added value’ under Horizon 2020 is the creation of transnational and multidisciplinary networks. The programme offers unique opportunities for collaboration and networking between a critical mass of researchers and innovators, generating knowledge and spillovers while finding joint solutions to global challenges. By far the lion’s share of the Horizon 2020 budget is spent on collaborative R&I projects. To fully reap the benefits across countries, networks need to remain open and easily accessible to new participants. This requires a good understanding of how researchers work together under the programme.

8 The UK became a non-EU country on 1 February 2020, but the Withdrawal Agreement provides that UK-based legal entities will continue to be fully eligible to participate in and receive funding from Horizon 2020, as if the UK were a Member State, until the closure of this programme. 9 Most Horizon 2020 projects are implemented by consortia of partners from different countries. This generates ‘European added value’ that goes beyond individual countries, but this is not captured by data on applicants’/participants’ country of origin. 10 i.e. countries that joined the EU in 2004 and subsequently (Bulgaria, Croatia, Cyprus, Czechia, Estonia, Hungary, Latvia, Lithuania, Malta, Poland, Romania, Slovakia and Slovenia); the ‘EU-15’ are the countries that were already Member States. 11 https://ec.europa.eu/info/research-and-innovation/strategy/support-policy-making/support-national-research-and-innovation-policy- making/srip-report_en

5 Monitoring flash 2 provides a dynamic analysis of the networks of participation in the framework programmes in the past decade. It shows the massive scale of the network, with over 1.5 million one-to-one collaborations between organisations under Horizon 2020 and 5 million collaborations generated since FP6. It also finds that the network is highly inter-connected (on average, there are only three ‘degrees of separation’ between participants), is very dynamic over time and seems to have been opening up to less well-connected participants since FP6, which might signal the entry of smaller players. While the number of collaborations involving EU-13 participants clearly fell between FP6 and FP7, a growing proportion of multi-partner Horizon 2020 projects involve at least one EU-13 participant. It seems that participants’ geographical and cultural proximity has an important influence on the structure of the network.

The analysis also indicates that, while the most connected countries in the network are the largest ones, some countries ‘punch above their weight’. If account is taken of countries’ size, the most central country is Finland, followed by Slovenia. Overall, Slovenia, Cyprus, Estonia and Malta are as central as the EU-15 countries. Still, several EU-13 countries are consistently low down in the ranking. The positions of the UK and Hungary have dropped significantly between FP7 and Horizon 2020. Overall, there is scope for improving several countries’ connectivity and centrality in the participation networks, especially those with lower R&I performance.

Monitoring flash 4 digs deeper into the analysis of international cooperation under the programme. It is essential that researchers and innovators in the EU have access to knowledge, expertise and facilities elsewhere. International collaboration is needed to tackle societal challenges that are global by nature and it is key to ensuring that EU companies stay globally competitive. Also, EU-level action can help to shape worldwide, multilateral R&I policy agendas, activities and cooperation mechanisms. Horizon 2020 is an essential instrument through which the EU can implement its strategy of international cooperation on the basis of common priorities and mutual benefits, taking account of scientific and technological capacities, market opportunities and expected impact.

The analysis shows a degree of variation in patterns of cooperation with non-EU countries; this is a reflection of Member States’ strategic targeting and pursuit of diverse objectives and benefits. Most collaborations are with a group of countries with advanced R&I capabilities, in particular through researcher mobility schemes (e.g. MSCAs), but also through specific projects and multilateral initiatives to support sustainable development and address global societal challenges, inter alia with developing economies. Nations with strong R&I performance, such as Switzerland, Norway and Israel, are the most active associated countries, while almost a third of participations from non-associated countries are from the United States (partly due to large US participation in MSCAs).

Horizon 2020 also contributes to the integration of R&I systems in the ERA for non-EU countries with a relative lack of R&I capacity, including through researcher mobility. Participation from such countries is still challenging, as reflected in their under-EU average performance in terms of the quantity, quality and success of applications.

Finally, the analysis highlights the influence of international cooperation on research quality: peer-reviewed FP7 and Horizon 2020 publications involving a contributor from at least one associated or other non-EU country are cited more than Member State-only publications and at least three times more than the world average.

6 Horizon 2020 is addressing global challenges and the SDGs

Monitoring flash 5 on the sustainable development goals (SDGs)12 maps Horizon 2020’s contribution to the United Nations’ global political agenda addressing a range of pressing social, economic and environmental challenges. The programme is one of the means by which the Commission is steering the requisite transitions, by investing in new knowledge and solutions that will help Europe to become the first climate-neutral continent in the world by 2050, grasp the opportunities of the digital age and develop an economy that works for and protects its people.

The analysis shows that up to 84% of current Horizon 2020 investments relate to at least one of the SDGs. Overall, this represents close to €40 billion invested in over 20 000 different projects, carried out by nearly 30 000 different beneficiaries from 152 countries. All three pillars of Horizon 2020 appear to contribute to the SDGs to a similar degree (top-down and bottom-up investment).

All SDGs are equally important for a sustainable future and there are many potential ‘sustainability pathways’. The biggest proportion of the investment relates to climate action and ‘good health and wellbeing’. The EU’s R&I investment in ‘responsible production and consumption’ seems low, especially given its current performance gap when it comes to achieving the SDG targets in this area.

Sustainability is complex and genuine transformation requires a shift away from conventional silos, sectors and disciplines to more interconnected and systemic thinking. The SDG agenda explicitly recognises the linkages between the different SDGs and the existence of both trade-offs and synergies. Accordingly, Horizon 2020 investments are also highly interconnected: on average, a Horizon 2020 project could relate to three different SDGs. The key challenge lies in making the right policy choices, so as to leverage the synergies and minimise the potential trade-offs.

Horizon 2020 is delivering value in key strategic areas underpinning the future of the EU economy

Monitoring flash 4 focuses on patents and the patenting activity of innovators benefiting from FP7 and Horizon 2020, based on self-reported project results.

On the basis of the available data, the analysis identifies 2 776 inventions (patent families13) self-reported by FP7 and Horizon 2020 beneficiaries; this represents 10 920 patents awarded worldwide. Overall, 75% of all patents are owned by European entities, over half (52%) of which are SMEs. The latter proportion is much bigger than relative SME representation in the programme (22%), but much lower than the percentages of patents owned by SMEs worldwide and in the EU. Figures confirm that the framework programmes develop technologies and innovations that underpin the European economy and are more interdisciplinary than average. Most self-reported inventions under the programmes relate to the health sector (in areas such as biotechnology, pharmaceuticals and organic chemistry) and a limited number relate to environmental technologies. However, this may reflect the policy priorities under FP7; a different picture is likely to emerge as a result of the explicit focus on climate action under Horizon 2020 and Horizon Europe.

Inventions arising from the framework programmes are protected by 61 different patent offices worldwide, but the main target markets are in Europe (75% of the inventions are protected by the European Patent Office and 74% in Member States) followed by the USA (where 74% of the inventions are protected). 28% of the inventions are protected in China. On average, each

12 The 17 SDGs were adopted by 193 UN member states in 2015. 13 i.e. sets of patents that protect the same invention disclosed by the same inventor in different countries. The number of ‘patent families’ indicates the number of distinct inventions protected.

7 invention is protected in 3.7 different markets. In general, these patents are associated with an above-average estimated market value. Patents registered in the USA are normally valued higher than those registered in Europe, but patents for inventions from the framework programmes achieve the highest valuations on Asian markets.

by Nelly Bruno and Martina Kadunc

FROM HORIZON 2020 TO HORIZON EUROPE MONITORING FLASH COUNTRY PARTICIPATION

August 2018

This Monitoring Flash is based on monitoring data of Horizon 2020 – the European Framework Programme for Research and Innovation 2014-2020 - and its predecessor, the Seventh Framework Programme (FP7). Widening participation is monitored regularly as a cross-cutting issue across Horizon 2020. This analysis covers the applications and participations from entities located in the different participating countries for the first 4.5 years of Horizon 2020 implementation. However, most Horizon 2020 projects are implemented by consortia of partners from different countries, and they generate a European added value that goes beyond each country. This European added value is not captured by looking specifically at the country of origin of the applicants and participants. HORIZON 2020 - 4.5 YEARS OF IMPLEMENTATION Key overview data €33.1b 19 292 11.9% 88 374 27 355 of EC contribution grants signed, from of proposals are participations in signed distinct participants allocated to signed grants 155 196 proposals successful grants from 148 countries

COUNTRY PARTICIPATION IN HORIZON 2020 Evolving heterogeneity

. Compared to the previous Framework Programme (FP7) the share of EU funding going to ‘EU13’1 countries has slowly increased (from 4.2% to 4.8% in Horizon 2020). This is proportionate to their share in the EU wide investments in research and development (R&D) (4.4%). The share of applications from EU13 entities has also slightly increased (from 9.6% in FP7 to 10% in Horizon 2020) but this remains relatively low compared to their share of the EU’s scientists and engineers (17%). There are also indications that an increasing share of Horizon 2020 multi-beneficiary projects are involving at least one EU13 participant, reversing a downward trend observed under FP7. . Different country groupings conceal noticeable performance differences among Member States and across Horizon 2020 programme parts. Some EU13 countries perform better than some EU15 countries in the 2018 European Innovation Scoreboard; and/or have a relatively high number of applications compared to their population of scientists and engineers. At the same time, some EU15 countries score poorly in the Scoreboard and/or have a relatively low participation in Horizon 2020. Data overall still shows that the more a country invests nationally in its R&D capacity, the more funding from the Programme it receives.

1 EU13 Member States are meant as Bulgaria; Croatia; Cyprus; Czech Republic; Estonia; Hungary; Latvia; Lithuania; Malta; Poland; Romania; Slovakia; and Slovenia, whereas EU15 countries are the other 15 Member States of the European Union.

Introduction

This is the first in a series of Monitoring Flash reports prepared by the Directorate-General for Research and Innovation (DG RTD) to provide up-to-date evidence on the implementation of Horizon 2020 – the European Framework Programme for Research and Innovation. This evidence base should inform policy discussions on the Commission proposal for the successor programme, Horizon Europe (2021-2027). Data covers the first 4.5 years of implementation of Horizon 2020, and the full implementation of FP7.2 Detailed data tables per country and programme parts are available in the Data Annexes3.

Applications and Success Rates

Evidence shows that stakeholders from all EU Member States are engaged more closely with Horizon 2020 compared to the previous Framework Programme (FP7), with the number of applications submitted having already surpassed the number submitted during the whole of FP7. Under Horizon 2020 to date, more than 530,000 applications within 155,000 projects’ proposals have been submitted, from entities located in more than 200 countries. This represents almost double the number of applications to FP7 per year.

Slightly more than half (53%) of these applications involve entities located in five countries: United Kingdom, Italy, Germany, Spain and France. Overall, almost eight out of ten applications comes from entities in EU15 countries. These countries also constitute 83% of the population of scientists and engineers in Europe and represent 96% of EU28 investments in research and development (R&D).

Entities located in countries that are ‘Innovation leaders’ or ‘Strong innovators’ according to the grouping of the European Innovation Scoreboard 2018 tend to also apply more frequently than the other country groups. Some 59% of the applications from the EU28 countries come from the 12 countries ranked as ‘Innovation leaders’ or ‘Strong innovators’, while 41% from the 16 countries ranked as ‘Moderate’ or ‘Modest’ innovators.

Representing 4% of EU28 national investments in R&D and 17% of European scientists and engineers4, the so- called EU13 countries account for 10% of applications, with a slight increase compared to FP7 (9.6%). Notably Cyprus, Slovenia, Malta and Estonia are among the top 5 European countries – together with Greece - when looking at the number of applications according to their population of scientists and engineers. On the other hand, the relatively high number of scientists and engineers in Poland are typically involved in very few proposals, three times less than the rest of Europe (8.5 applications on average against 27 for EU28). On average, each applicant to Horizon 2020 submits 6 different proposals, as compared to 11 over the whole duration of FP7. However applicants from EU13 countries tend to apply less frequently than the ones from EU15 countries (3.8 against 6.5 applications per distinct applicant).

2 This Monitoring Flash includes all fully evaluated calls, including from the Work Programmes of the Public-Private Partnerships (Joint Undertakings), but excludes Public-Public Partnerships, the EIT’s Knowledge and Innovation Communities (KICs) and direct actions of the Joint Research Centre. The data is stored in the Common Research Data Warehouse (CORDA), an internal database maintained by DG RTD. 3 For additional information and the latest implementation data, please check the Horizon 2020 Dashboard https://webgate.ec.europa.eu/dashboard 4 Scientists and engineers as defined by Eurostat (data table hrst_st_ncat)

Figure 1 Number of applications to FP7 and Horizon 2020 per year per Member State

14000

12000

10000

8000

6000

4000

2000

0 UK IT DE ES FR NL BE EL SE AT PT DK FI PL IE HU SI CZ RO BG CY EE HR SK LT LV LU MT FP7 9849 8029 9412 7430 6301 4158 2922 2865 2518 2037 1638 1501 1649 1583 1197 1000 746 920 927 552 403 330 314 355 279 193 141 133 Horizon 2020 13979132151301012760 9107 6568 4464 3778 3412 2903 2842 2707 2556 2266 1934 1411 1286 1257 1206 789 668 624 600 553 480 421 324 186 Difference FP7/ 42% 65% 38% 72% 45% 58% 53% 32% 36% 43% 73% 80% 55% 43% 61% 41% 72% 37% 30% 43% 66% 89% 91% 56% 72% 118% 130% 40% Horizon 2020

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, Cut-off date 1 July 2018

All country groups witnessed a strong drop in the success rates of applications between FP7 and Horizon 2020, from 21.7% to 14.7% for the programme as a whole. Countries with a high level of national expenditure for research and development compared to their population also show the highest success rates. Applications from entities in Belgium, France, Austria, the Netherlands, and Germany are the most successful, with close to 1 out of 6 applications being successful. Applications from entities in Bulgaria, Slovenia, Hungary, Croatia and Lithuania are the least successful. Success rates and other key data on applications are presented per country in the Data Annex, Table 1. The highest success rates can be found in the groups of ‘Innovation leaders’ and ‘Strong innovators’ with the notable exception of Finland and Slovenia, which display lower than average success rates.

Figure 2 Success rates of applications to Horizon 2020 and FP7 per country groups of applicants (EIS 2018, EU15/13)

25% 25% 20% 20% 15% 15% 10% 10% 5% 5% 0% 0% STRONG MODERATE MODEST Associated Third LEADER NOT EU Overall EU15 EU13 Overall INNOVATOR INNOVATOR INNOVATOR Countries Countries FP7 23,2% 24,0% 18,6% 15,3% 22,6% 21,7% FP7 22,1% 18,0% 21,7% 23,8% 21,7% Horizon 2020 15,2% 16,5% 12,8% 10,9% 15,5% 14,7% Horizon 2020 14,9% 11,8% 14,4% 17,8% 14,7% Difference FP7/ Difference FP7/ -34,5% -31,2% -31,1% -28,4% -31,7% -32,5% -32,2% -34,5% -33,8% -25,3% -32,5% Horizon 2020 Horizon 2020

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, Cut-off date 1 July 2018

Horizon 2020 Evaluators

All Horizon 2020 proposals are evaluated by independent experts. More than 22,000 different independent experts from 114 countries evaluated the 155,000 proposals submitted to date. In total 850,000 evaluations were carried out – on average each proposal is evaluated by at least 5 different experts.

72% (15,764) of evaluators were from EU15 countries and they carried out 66% of all evaluations during the first four and a half years of Horizon 2020. 16% (3454) of evaluators were from EU13 and they carried out 16% of all evaluations. The share of EU13 evaluators was proportionally much higher than the share of EU13 applications (10% of all applications). The number and share of evaluations carried out by independent experts per country is presented in the Figure 3 below.

Figure 3 Number and share of evaluations of Horizon 2020 proposals per country of evaluator

180000 18% 160000

140000

120000

100000 10% 9% 80000 8% 7% 7% 60000 4% 40000 4% 4% 3% 3% 2% 2% 2% 2% 2% 2% 1% 1% 20000 1% 1% 1% 1% 1% 1% 0% 0% 0% 0% 0 UK IT DE FR ES EL PT PL NL BE SE AT FI IE RO HU DK BG CZ SI HR SK LT CY EE MT LV LU Non EU

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, Cut-off date 1 July 2018

Participation Patterns and EU Investment

The overall participation patterns and EU investment from the Framework Programmes are largely correlated with the size of the national research and innovation systems, which are characterised by different levels of national investments in research and development and varying populations of scientists and engineers.

Analysis per different types of country grouping beyond the factor of size does not show clear trends – both the EU15, EU13 grouping and the country grouping based on the latest European Innovation Scoreboard ranking from 2018 indicate significant differences when looking at individual countries within each group.

More than 27,000 distinct beneficiaries received grants from Horizon 2020 during its first four and a half years. On average each beneficiary participated in the programme more than three times, hence the total number of participations in the programme being 88,374. Comparatively, more than 30,000 distinct beneficiaries received grants from FP7, while on average each beneficiary participated in the FP7 programme more than four times throughout its seven year lifetime.

Figure 4 compares participations per year by EU Member State in Horizon 2020 and FP7. The overall number of participations per year from beneficiaries located in EU Member States in Horizon 2020 remains similar to FP7 (an increase of 0.4% or 60 participations per year). More than 50% of all participations in both FP7 and Horizon 2020 were from beneficiaries located in Germany, the United Kingdom, Spain, France and Italy. These are also the EU Member States with the biggest national research and innovation systems (i.e. they are home to more than 60% of all scientists and engineers in Europe and more than 70% of all R&D investment in EU). Analysis shows that there is a strong correlation between the share of participations in the EU Framework Programme and the size of the national scientific and engineering workforce5; countries with a larger share of scientists and engineers have a larger share of participations.

Differences are observed when comparing the number of participations of entities from individual Member States between FP7 and Horizon 2020. Some Member States record a high increase in participations compared to FP7: Luxembourg (+64.1%), Cyprus (+ 45.7%) and Croatia (+45.4 %). Other Member States record a strong decrease in participations from FP7 to Horizon 2020: Hungary (-28.3%), the United Kingdom (-11.6%) and Germany (-11.2%).

5 The Pearson correlation coefficient is larger than 0.8.

Figure 4 Number of participations to Horizon 2020 and FP7 per year per Member State of beneficiaries

3.000 2.500 2.000 1.500 1.000 500 0 AT BE BG CY CZ DE DK EE EL ES FI FR HR HU IE IT LT LU LV MT NL PL PT RO SE SI SK UK FP7 519 857 102 67 207 2.66 409 80 541 1.66 417 1.86 59 237 287 1.74 61 37 48 29 1.21 319 349 156 659 134 71 2.60 Horizon2020 547 867 91 95 200 2.36 443 92 554 2.03 387 1.89 86 169 320 1.87 65 60 57 28 1.20 305 428 169 596 163 82 2.29 Difference % 5% 1% -11% 42% -4% -11% 8% 15% 2% 22% -7% 2% 45% -28% 12% 7% 7% 64% 17% -2% 0% -4% 23% 8% -10% 21% 15% -12%

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, Cut-off date 1 July 2018

Looking at different country groupings, the participations of non-EU countries in Horizon 2020 decreased by 9% when compared to FP7. Conversely, participations of both EU15 and EU13 countries slightly increased (0.2% for EU15 and 2% for EU13). However as seen in Figure 5, there are considerable differences within each group of countries. Based on the European Innovation Scoreboard ranking of 2018, participations from EU ‘Innovation ‘Leaders’ and ‘Strong Innovators’ decreased from FP7 to Horizon 2020, whereas participations from ‘Moderate’ and ‘Modest innovators’ increased (see Figure 5 below).

Figure 5 Number of participations to Horizon 2020 and FP7 per year per country groups (EIS 2018, EU15/13)

7.000 18.000 6.000 16.000 14.000 5.000 12.000 4.000 10.000 3.000 8.000 6.000 2.000 4.000 1.000 2.000 0 0 STRONG MODERATE MODEST Associated Third LEADER NOT EU EU15 EU13 INNOVATOR INNOVATOR INNOVATOR Countries Countries FP7 5.338 6.332 5.484 258 2.376 FP7 15.843 1.569 1.629 717 Horizon2020 4.992 6.156 6.066 260 2.166 Horizon2020 15.872 1.601 1.408 758 Difference % -6% -3% 11% 1% -9% Difference % 0,2% 2% -14% 6%

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, Cut-off date 1 July 2018

More than EUR 33.1 billion were invested through Horizon 2020 during its first four and a half years, or on average EUR 7.4 billion per year. This is a 13% increase from the FP7 average annual investment of EUR 6.5 billion.

Figure 6 compares the EU investment per year by countries of beneficiaries for EU Member States in Horizon 2020 and FP7. Given the overall increase in the budget of Horizon 2020, entities from all EU Member States except Bulgaria and Croatia were awarded a higher amount per year. In particular, entities located in Luxembourg, Cyprus, Estonia and Slovenia saw a doubling or near-doubling of their annual EU funding from Horizon 2020 compared to FP7

Figure 6 Annual EU investment through Horizon 2020 and FP7 (EUR million), by Member State of beneficiaries 1.400

1.200

1.000

800

600

400

200

0 AT BE BG CY CZ DE DK EE EL ES FI FR HR HU IE IT LT LU LV MT NL PL PT RO SE SI SK UK FP7 170 277 14 13 41 1.02 154 14 146 471 125 744 13 41 90 519 7 9 7 3 489 63 75 19 250 24 11 1.00 Horizon2020 210 350 14 28 46 1.17 187 27 167 662 163 776 13 46 126 608 9 19 11 5 552 68 119 25 258 43 19 1.05 Difference % 23% 26% -1% 114% 11% 15% 21% 99% 15% 41% 31% 4% 0% 11% 40% 17% 27% 120% 56% 49% 13% 8% 58% 30% 3% 75% 71% 5%

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, Cut-off date 1 July 2018

Looking at different country groupings (Figure 7) entities located in EU13 countries recorded on average a 30% increase in EU funding from Horizon 2020 per year when compared to FP7. At the same time, EU funding to third country entities decreased by more than half. Based on the European Innovation Scoreboard ranking of 2018, the annual EU funding to entities from the group of countries ranked as ‘Moderate innovators’ increased the most.

Figure 7 Annual EU investment through Horizon 2020 and FP7 (EUR million), by country group of beneficiaries

3.000 7.000 2.500 6.000 2.000 5.000 4.000 1.500 3.000 1.000 2.000 500 1.000 0 0 STRONG MODERATE MODEST Associated LEADER NOT EU EU15 EU13 Third Countries INNOVATOR INNOVATOR INNOVATOR Countries FP7 2.029 2.327 1.425 34 674 FP7 5.543 272 584 86 Horizon2020 2.231 2.678 1.828 39 585 Horizon2020 6.422 354 548 37 Difference % 10% 15% 28% 17% -13% Difference % 16% 30% -6% -57%

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, Cut-off date 1 July 2018

In terms of the overall distribution of Horizon 2020 funding to beneficiaries across the EU28 investment patterns closely follow the participation patterns observed above: more than 50% of both Horizon 2020 and FP7 funds went to entities located in Germany, the United Kingdom, Spain, France and Italy. Again, these are also the Member States with the biggest national research and innovation systems. As in the case of the number of participations, the analysis shows a strong correlation6 between the share of funding from the EU Framework Programme and the size of the national research and development investments; beneficiaries from countries with higher national investment in research and development (GERD) obtain a larger share of EU funding (see also Figure 15).

6 The Pearson correlation coefficient is larger than 0.8.

To account for such differences in the sizes of the national research and innovation investments, Figure 8 compares the annual Horizon 2020 investment to the national expenditures in R&D (GERD). On average Horizon 2020 invests EUR 0.20 for each EUR of national R&D investment in EU Member States. Horizon 2020 funding awarded to beneficiaries from Cyprus is much higher in relative terms (EUR 0.30 of Horizon 2020 funding for each EUR of national R&D investment). Similarly, beneficiaries from Estonia, Latvia and Greece receive EUR 10 cents of investment from Horizon 2020 for each EUR invested nationally. Countries categorised as Innovation ‘Leaders’ in the European Innovation Scoreboard 2018 attract 30% of EU investments through Horizon 2020 when they represent 28% of the EU28 investments in R&D.

Figure 8 Annual EU investment through Horizon 2020 per EUR of national gross expenditures in R&D (GERD) by country of beneficiary

0,35 0,31 0,30

0,25

0,20

0,15 0,10 0,10 0,10 0,10 0,07 0,05 0,05 0,05 0,04 0,04 0,04 0,05 0,03 0,03 0,03 0,03 0,03 0,03 0,03 0,03 0,03 0,03 0,02 0,02 0,02 0,02 0,02 0,02 0,01 0,00 CY EE LV EL MT SI PT ES IE NL BG HU HR BE RO SK LU LT IT FI UK DK AT SE PL FR CZ DE

Horizon 2020 investment per GERD Average EU-28

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, Cut-off date 1 July 2018

Other Indicators of Participation Patterns

The number of projects with at least one participant from each country and country group, as well as other indicators relevant to monitor participation patterns and EU investment in the Framework Programme are included in the Data Annex (Table 2 and Table 6). These generally follow similar patterns described in the analysis above.

When focusing specifically on the EU13 country grouping and looking at the openness of networks, in FP7 21% of all projects had at least one EU13 participant whereas this represents 17% of projects in Horizon 2020. This decrease is largely explained by the increase in the number of mono-beneficiary projects in Horizon 2020 (61%) compared to FP7 (52%)7.

Figure 9 Share of projects with at least one EU13 participant in Horizon 2020 and FP7 in total number of projects

FP7 (25,782 projects) 2,1% 49,4% 19,1% 29,4%

Horizon 2020 (19,292 projects) 2,6% 58,7% 14,0% 24,6%

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

EU-13 mono-beneficiary Other mono-beneficiary EU-13 multi-beneficiary Other multi-beneficiary

7 NB: The budget share of multi-beneficiary projects in Horizon 2020 is 72%.

To discount for the effect of the larger share of mono-beneficiary projects, EU13 participation is further analysed for multi-beneficiary projects only. On average, 40% of all multi-beneficiary projects in FP7 had at least one EU13 participant compared to 36% in Horizon 2020. Figure 10 depicts a decreasing annual trend as regards the share of multi-beneficiary projects with at least one EU13 participant under FP7: from 49% at the beginning of the programme (2007) to 32% at its end (2013). In Horizon 2020, the trend appears to reverse towards more open networks to EU13 participants: 40% of multi-beneficiary projects funded in 2017 had at least one EU13 participant.

Figure 10 Share of multi-beneficiary projects with at least one EU13 participant in Horizon 2020 and FP7 in total number of projects

60%

50%

40%

30%

20% one one EU13 participant 10%

Share of collaborative projects with at least least at with projects collaborative of Share 0% 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017

FP7 Horizon 2020

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, cut-off data 1 July 2018

When looking at the different activities within Horizon 2020, the country participation patterns remain similar to those observed for the programme as a whole. However, three actions show some differences: the SME Instrument, the Coordination and Support Actions and the ERA-Net Cofund (see Data Annex, Table 3).

Scientific and Innovation Performance

The scientific and innovation performance of countries is influenced by a multitude of factors wider than participation in the Framework Programme. Looking at the level of participation in Horizon 2020 weighted by the number of scientists and engineers in the country, only those countries with a very low level of participation such as Bulgaria, Romania and Poland also underperform in terms of scientific and innovation outputs (as measured by typical indicators on peer-reviewed (co)publications, their citations and patents). The other countries display a more diverse picture as depicted in Figure 11.

Figure 11 Horizon 2020 participations per country of beneficiary compared to overall performance of EU countries in key scientific and innovation performance indicators

SCIENTIFIC PUBLICATIONS NR OF EMPLOYMEN AMONG THE TOP INTERNATIONA PUBLIC- PARTICIPATIONS PCT PATENT T IN 10% MOST CITED L SCIENTIFIC PRIVATE CO- EUROPEAN IN SIGNED APPLICATION KNOWLEDGE PUBLICATIONS CO- PUBLICATION INNOVATION COUNTR GRANTS PER S PER -INTENSIVE EU15/ EU13 WORLDWIDE AS % PUBLICATIONS S PER SCOREBOAR Y CODE THOUSANDS BILLION GDP ACTIVITIES GROUPING OF TOTAL PER MILLION MILLION D COUNTRY SCIENTISTS AND (IN PPS), (% OF TOTAL SCIENTIFIC POPULATION, POPULATION, GROUP 2018 ENGINEERS IN 2015 EMPLOYMEN PUBLICATIONS OF 2017 2017 THE COUNTRY T, 2017 THE COUNTRY, 2015 CY 15,5 9,0 1283,3 21,1 0,8 17,0 MODERATE EU13 LU 10,6 13,1 1715,0 25,4 1,8 22,0 LEADER EU15 MT 9,8 10,7 597,4 0,0 1,3 18,4 MODERATE EU13 EL 9,7 9,0 608,3 10,5 0,5 12,1 MODERATE EU15 SI 9,6 8,6 1134,6 56,1 1,6 13,7 STRONG EU13 AT 8,7 11,1 1375,8 82,3 4,7 15,0 STRONG EU15 EE 8,5 8,2 1077,8 10,6 1,0 13,5 MODERATE EU13 BE 8,0 12,6 1467,6 80,0 3,2 15,6 STRONG EU15 IT 7,9 10,4 631,9 22,2 2,2 13,7 MODERATE EU15 DK 7,0 13,4 2345,9 162,8 6,1 15,1 LEADER EU15 ES 6,6 9,3 732,1 21,1 1,4 12,5 MODERATE EU15 NL 6,6 14,6 1628,1 99,3 5,8 17,1 LEADER EU15 FI 6,3 10,8 1658,8 85,4 7,4 16,2 LEADER EU15 IE 6,2 12,6 1249,3 45,4 1,8 20,6 STRONG EU15 PT 5,4 9,0 918,9 13,2 0,9 10,6 MODERATE EU15 LV 4,9 6,2 315,4 1,0 0,8 12,1 MODERATE EU13 FR 4,8 11,0 726,2 42,8 4,0 14,5 STRONG EU15 SE 4,6 12,1 2018,8 130,6 9,1 18,5 LEADER EU15 HR 4,3 4,6 492,3 17,3 0,6 11,6 MODERATE EU13 SK 3,7 6,2 438,8 10,3 0,5 10,6 MODERATE EU13 DE 3,3 11,3 812,2 62,4 6,1 14,8 STRONG EU15 HU 3,1 6,9 456,3 29,6 1,3 11,6 MODERATE EU13 UK 3,0 15,0 1222,3 65,1 3,1 18,5 LEADER EU15 LT 2,9 4,3 450,5 3,9 0,8 9,7 MODERATE EU13 CZ 2,7 6,6 754,8 21,0 0,9 12,9 MODERATE EU13 BG 2,1 4,2 226,6 3,0 0,6 10,2 MODEST EU13 RO 1,4 4,8 181,8 3,7 0,2 7,7 MODEST EU13 PL 1,1 5,1 296,6 5,4 0,7 10,3 MODERATE EU13 EU28 n.a 10,57 517,45 40,93 3,53 14,20 Source: European Commission, DG RTD based on CORDA data, and data from the European Innovation Scoreboard 2018 (incl. Eurostat data)

Looking specifically at the quality and influence of the outputs produced with Horizon 2020 funding so far, the interim evaluation of Horizon 20208 provided indications of the high quality and reputation of the research and innovation activities performed in the Framework Programmes.

The preliminary assessment of the Field Weighted Citation Index (FWCI)9 of the 4043 Horizon 2020 peer- reviewed publications, carried out at the time of the interim evaluation in 2017, confirmed the trends observed in the period 2007-2013 for FP7: publications from FP7 and Horizon 2020 projects are cited at more than twice the world average (FWCI of 2.46). For 2015 and 2016, the EU28 countries’ Horizon 2020-funded output was represented 3.74 times more in the world’s top 1 % of cited research than the overall EU28 publication output.

8 European Commission, Interim evaluation of Horizon 2020, 2017, https://ec.europa.eu/research/evaluations/pdf/book_interim_evaluation_horizon_2020.pdf 9 Field-weighted Citation Impact normalises citation differences between research fields, with a world average set to 1.0.

Looking at the country groups of EU15 and EU13, Horizon 2020-funded publications were proportionally higher in the top 1% category by factors of 3.65 and 5.57, respectively. The EU13 group enjoyed the highest relative increase, between 2015 and 2016, over their own overall FWCI, with 1.84 and 2.29 ratio increases, respectively.

Figure 12 Field Weighted Citation Index for FP7 publications (left) and for Horizon 2020 (right) FP7 Horizon 2020

4 4

3 3

2 2

1 1

0 0 2007-2013 2015 2016

EU28 Associated Countries Third Countries World Average: 1.0

Based on ECDG, Elsevier, Overall output of selected geographical group comparators and related FP7- and H2020-funded publication output, 2017

Overall, the share of EU13 publications in the world top 1% most cited increased from 3.5% in FP7 to 4.4% in Horizon 2020. FP publications are at least 3 times more likely to belong to the top 1% compared to non-FP publications.

Figure 13 Share of FP7 and Horizon 2020 publications in world top 1% (2007-2016) Share of FP7 and Horizon 2020 publications in world top 1% (2007 - 2016)

6,2% 5,3% 5,4% 5,3% 4,3% 4,4% 4,4% 3,8% 3,9% 3,5%

1,1% 1,2% 1,2% 1,1% 1,2% 1,2% 0,8% 0,9% 0,8% 0,9%

E28 E15 E13 AC TC E28 E15 E13 AC TC

FP non-FP

Source: Based on ECDG and Elsevier, Overall output of selected geographical group comparators and related FP7- and H2020-funded publication output, 2017

In terms of interdisciplinarity, the share of Horizon 2020 publications which are interdisciplinary is relatively high and increasing slightly compared to FP7. For the EU28, out of their total number of Horizon 2020 publications, 7.55% are interdisciplinary (compared to 7.45 % in the first three years of FP7). The EU15 share is 7.29% (compared to 7.53% in the first three years of FP7). For the EU13, the share is 10.19% (compared to 5.87% in the first three years of FP7). This means that the EU13 produces more interdisciplinary publications when compared to the EU15, and that the share of such publications among the EU13 countries in Horizon 2020 has doubled compared to what it was in FP7. So far, when looking only at interdisciplinary Horizon 2020-funded research, the FWCI indicates that these Horizon 2020 interdisciplinary publications are cited 78% more than the world average in this field (FWCI of 1.78). This is rising on a yearly basis.

In terms of international collaborations, 77% of publications from EU13 authors in FP7 were international. In Horizon 2020, the share decreased slightly to 72% in 2015 and 74% in 2016.

Figure 14 Share of international collaborations in FP7 and Horizon 2020 publications per country group (2007-2016)

* Disclaimer: international collaboration is defined as those publications in which at least one co-author comes from outside the geographical group. Source: Based on ECDG and Elsevier, Overall output of select geographical group comparators and related FP7- and H2020-funded publication output, 2017

According to an external study based on counter-factual analysis, EU-funded research teams are around 40% more likely to be granted patents or produce patent applications than non-funded units. The data also shows that the patents produced under the Framework Programmes are of higher quality and likely commercial value than similar patents produced elsewhere10.

Overall messages for Horizon Europe

Figure 15 compares the Horizon 2020 investment per country of beneficiary with the corresponding national investments in research and development: the more a country invests nationally in its R&D capacity, the more funding from the Framework Programme it receives11. Figure 15 Horizon 2020 investment per country of beneficiary, in signed grants, compared to the national investments in research and development (Intramural R&D expenditures - GERD) per country, in EUR per inhabitant

Source DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, cut-off data 1 July 2018, and Eurostat data on GERD (2016)

A set of dedicated measures exist under Horizon 2020 to help spread excellence and widen participation such as Teaming, Twinning and ERA-Chairs (EUR 328 million invested so far, including 73% in Widening countries, see

10 PPMI, Assessment of the Union Added Value and the Economic Impact of the EU Framework Programmes (FP7, Horizon 2020), 2017 11 The Pearson correlation coefficient is larger than 0.8.

details in Data Annex, Table 8). Initiatives such as COST (promoting and networking pockets of excellence) and the Policy Support Facility12 (providing on-demand advice to policy makers on national research and innovation systems) also provide support to low-performing countries. As stated in the interim evaluation of Horizon 202013, the main expected outputs from these measures are related to the strengthened institutional, scientific and networking capacities of centres of excellence and knowledge and research institutions located in low- performing regions and Member States – on the basis of partnerships with internationally leading institutions and researchers – improved research and innovation policy frameworks, and support provided to strategic planning and implementation.

Overall, evidence shows the share of funding going to the EU13 countries has a slowly increasing trend under Horizon 2020 compared to FP7, which is proportionnate to their share in the overall investments of all EU Member States in research and development. There are also indications that an increasing share of multi- beneficiary Horizon 2020 projects are involving at least one EU13 participant, reversing a trend observed under FP7, even as the share of EU13 applications remains low relative to the number of scientists and engineers in their population.

The evidence presented in this Flash raises the question whether an EU13 versus EU15 framing of issues around spreading excellence and widening participation to the Framework Programme remains valid and useful in the future. Different country groupings conceal noticeable performance differences among Member States and across Horizon 2020 programme parts. The differences in research and innovation performance among Member States – sometimes called divide – are determined by a multitude of factors beyond the influence of the Framework Programme, such as the national priorities, the level of private and public investments (incl. through public procurement), the availability and quality of infrastructures, human capital and skills, the access to finance, the support measures in place, etc. requiring tailored policy mixes in Member states in line with each country’s specific challenges14.

Raising the level of participation in the Framework Programme of low research and innovation performing countries remains a complex issue which needs intervention at various levels. Widening participation in the EU Framework Programme is a shared responsibility to be addressed nationally and at EU level in a complementary and synergetic way whilst respecting the principles and role of each level and the measures used. The Framework Programme can stimulate reforms and leverage more and better investments across Europe if excellence is maintained as the programme's main driver and evaluation criterion. The search for excellence is what drives reforms, ensures value for money and delivery of measures that mutually benefit all parties involved and, ultimately, enables Europe to compete worldwide.

Towards Horizon Europe – based on the Commission proposal

The EU now needs to raise the bar in the quality and impact of its research and innovation system, requiring a revitalised European Research Area (ERA)15 , better supported by the EU's research and innovation Framework Programme. Specifically, a well-integrated yet tailored set of EU measures16 is needed, combined with reforms and performance enhancements at national level (to which the Smart Specialisation Strategies supported under the European Regional Development Fund can contribute) and, in turn, institutional changes within research funding

12 The Policy Support Facility (PSF) - launched under Horizon 2020 - works on a demand-driven basis and it offers, on a voluntary basis, high level expertise and tailor-made advice to national public authorities. Through its services, it has already been instrumental in provoking policy change in countries such as Poland, Bulgaria, Moldova or Ukraine and in bringing forward policy changes, driven by exchanges of good practice, in areas such as R&D tax incentives, open science, performance-based funding of public research organisations and the inter-operability of national research and innovation programmes. 13 European Commission, Interim evaluation of Horizon 2020, 2017 and Annex 2 Part Q for dedicated evaluation of Spreading Excellence and Widening Participation Programme part: https://ec.europa.eu/research/evaluations/index.cfm?pg=h2020evaluation 14 For a more in-depth analysis see for instance European Commission, Science, Research and Innovation Performance of the EU 2018; Strengthening the Foundations for Europe's Future, 2018 15 The ERA Progress Report of 2018 16 Council Conclusions on the ERA Roadmap, 19 May 2015

and performing organisations, including universities. By combining efforts at EU level, synergies can be exploited and the necessary scale can be found to make support to national policy reforms more efficient and impactful.

The activities supported under the proposed ‘Strengthening the European Research Area’ part of Horizon Europe addresses ERA policy priorities, while generally underpinning all parts of Horizon Europe. Activities may also be established to foster brain circulation across ERA through mobility of researchers and innovators.

Reducing disparities in research and innovation performance by sharing knowledge and expertise across the EU will help countries and regions that are lagging behind in terms of research and innovation performance, including the EU outermost regions, to attain a competitive position in the global value chains. Activities may also be established to foster brain circulation right across ERA and better exploitation of existing (and possibly jointly managed EU programmes) research infrastructures in the targeted countries through mobility of researchers and innovators.

A set of targeted measures include:

. Teaming, to create new centres of excellence or upgrade existing ones in eligible countries, building on partnerships between leading scientific institutions and partner institutions; . Twinning, to significantly strengthen a university or research organisation from an eligible country in a defined field, by linking it with internationally-leading research institutions from other Member States or Associated Countries. . ERA Chairs, to support universities or research organisations attract and maintain high quality human resources under the direction of an outstanding researcher and research manager (the 'ERA Chair holder'), and to implement structural changes to achieve excellence on a sustainable basis. . European Cooperation in Science and Technology (COST), involving ambitious conditions regarding the inclusion of eligible countries, and other measures to provide scientific networking, capacity building and career development support to researchers from these target countries. 80% of the total budget of COST will be devoted to actions fully aligned with the objectives of this intervention area. The above mentioned funding lines will facilitate specific research elements customised to the particular needs of the actions.

Policy reforms at national level will be mutually reinforced through the development of EU-level policy initiatives, research, networking, partnering, coordination, data collection and monitoring and evaluation. Low research and innovation performing countries will be able to also benefit from this general action. This includes:

. Support to national research and innovation policy reform, including though a strengthened set of services of the Policy Support Facility (PSF) (i.e. peer reviews, specific support activities, mutual learning exercises and the knowledge centre) to Member States and Associated Countries, operating in synergy with the European Regional Devleopment Fund, the Structural Reform Support Service (SRSS) and the Reform Delivery Tool; . Providing researchers with attractive career environments, skills and competences needed in the modern knowledge economy17. Linking the ERA and the European Higher Education Area by supporting the modernisation of universities and other research and innovation organisations, through recognition and reward mechanisms to spur actions at national level, as well as incentives promoting the adoption of open science practices, entrepreneurship (and links to innovation ecosystems), trans-disciplinarity, citizen engagement, international and inter-sectoral mobility, gender equality plans and comprehensive approaches to institutional changes. In that context, also complementing the Erasmus programme support for the European Universities initiative, in particular its research dimension, as part of developing new joint and integrated long term and sustainable strategies on education, research and innovation based on trans-disciplinary and cross-sectoral approaches to make the knowledge triangle a reality, providing impetus to economic growth.

17 Including notably the European Charter for researchers, the code of conduct for the recruitment of researchers, EURAXESS and RESAVER Pension Fund.

DATA ANNEXES Table 1 Applications to the Framework Programmes (FP7, Horizon 2020) and success rates, per country of applicants

Nr of FP7 Nr of Horizon Nr of FP7 Intramural Share of Nr of Horizon 2020 Nr of Horizon Nr of applications Success rate of Success rate of European European eligible % of total 2020 eligible R&D Scientists applications in 2020 per Thousands Horizon 2020 FP7 eligible Innovation Innovation EU15/ Country applications Horizon 2020 applications applications expenditure and Eligible Proposals applications Scientists and applications, applications, Scoreboard Scoreboard EU13 code per country applications, per distinct per distinct (GERD) - Engineers in per country of per country of Engineers in the per country of per country of Country Country Group of applicant per country applicant per applicant per Euro per EU28 by applicant applicant/ year population applicants applicants Group 2018 Scores 2018 per year country country inhabitant country

SE 15356 3412 2518 2,9% 26,4 15,6% 23,3% 6,0 13,1 1537 3,3% LEADER 0,71 EU15 DK 12183 2707 1501 2,3% 42,6 14,9% 24,0% 6,6 12,6 1396 1,6% LEADER 0,67 EU15 FI 11500 2556 1649 2,2% 41,8 13,5% 21,2% 5,5 13,0 1080 1,6% LEADER 0,65 EU15 NL 29554 6568 4158 5,6% 35,7 16,5% 25,5% 6,5 12,3 841,1 4,7% LEADER 0,65 EU15 UK 62905 13979 9849 11,8% 18,2 14,9% 22,6% 7,2 14,9 618,7 19,6% LEADER 0,61 EU15 LU 1460 324 141 0,3% 57,3 15,7% 18,3% 5,4 8,7 1143,9 0,1% LEADER 0,61 EU15 DE 58544 13010 9412 11,0% 17,9 16,3% 24,0% 6,3 12,3 1124,7 18,5% STRONG 0,60 EU15 BE 20088 4464 2922 3,8% 41,2 17,6% 26,1% 6,6 12,3 929,9 2,8% STRONG 0,59 EU15 IE 8701 1934 1197 1,6% 37,6 14,8% 21,6% 6,1 12,1 686,4 1,3% STRONG 0,58 EU15 AT 13063 2903 2037 2,5% 46,2 16,9% 22,1% 6,2 12,2 1255 1,6% STRONG 0,58 EU15 FR 40983 9107 6301 7,7% 22,9 17,2% 24,9% 6,2 12,4 750,4 10,2% STRONG 0,55 EU15 SI 5787 1286 746 1,1% 76,0 10,6% 15,7% 4,5 9,3 392 0,4% STRONG 0,47 EU13 CZ 5657 1257 920 1,1% 16,9 13,8% 20,5% 4,7 9,4 280,8 1,9% MODERATE 0,42 EU13 PT 12787 2842 1638 2,4% 35,7 12,8% 18,3% 6,2 10,9 227 2,0% MODERATE 0,41 EU15 MT 838 186 133 0,2% 65,5 13,4% 18,5% 4,4 8,8 140,1 0,1% MODERATE 0,40 EU13 ES 57422 12760 7430 10,8% 41,2 13,9% 19,2% 6,3 11,0 285,5 7,9% MODERATE 0,40 EU15 EE 2807 624 330 0,5% 57,9 13,0% 20,6% 4,3 10,0 205,4 0,3% MODERATE 0,40 EU13 CY 3008 668 403 0,6% 109,0 12,0% 14,8% 6,3 11,5 107,7 0,2% MODERATE 0,39 EU13 IT 59466 13215 8029 11,2% 55,6 12,2% 18,5% 5,3 11,9 356,2 6,1% MODERATE 0,37 EU15 LT 2162 480 279 0,4% 21,3 11,8% 20,0% 3,7 8,1 113,4 0,6% MODERATE 0,36 EU13 HU 6350 1411 1000 1,2% 25,5 10,7% 20,1% 3,6 9,6 139,5 1,4% MODERATE 0,33 EU13 EL 17001 3778 2865 3,2% 66,0 13,0% 16,6% 7,7 16,7 162,7 1,5% MODERATE 0,33 EU15 SK 2487 553 355 0,5% 24,6 12,7% 17,9% 3,3 7,1 118,1 0,6% MODERATE 0,32 EU13 LV 1896 421 193 0,4% 36,6 12,0% 22,0% 3,6 7,0 56,1 0,3% MODERATE 0,29 EU13 PL 10196 2266 1583 1,9% 8,5 12,2% 18,6% 3,8 10,5 108,3 6,8% MODERATE 0,27 EU13 HR 2699 600 314 0,5% 30,4 11,7% 17,3% 3,5 6,7 93,6 0,5% MODERATE 0,26 EU13 BG 3551 789 552 0,7% 17,9 9,5% 16,4% 2,7 7,5 52,5 1,1% MODEST 0,23 EU13 RO 5426 1206 927 1,0% 10,0 11,9% 14,6% 3,6 8,4 41,4 3,1% MODEST 0,16 EU13

EU13 52864 11748 7735 10,0% 17 11,8% 18,0% 3,8 9,0 n.a 17,2% n.a n.a n.a

EU15 421013 93558 61648 79,3% 29 14,9% 22,1% 6,3 12,6 n.a 82,8% n.a n.a n.a

EU28 473877 105306 69382 89,2% 27 14,6% 21,6% 5,9 12,0 n.a 100,0% n.a n.a n.a

Associated 39411 8758 6730 7,4% n.a 14,4% 21,7% 4,7 10,4 n.a n.a n.a n.a n.a Countries

Third 17912 3980 4278 3,4% n.a 17,8% 23,8% 3,1 6,3 n.a n.a n.a n.a n.a Countries

Total 531200 118044 80440 100% n.a 14,7% 21,7% 5,6 11,3 n.a n.a n.a n.a n.a Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, Cut-off date 1 July 2018, Eurostat data for GERD and HRST, European Innovation Scoreboard 2018

Table 2 Participations and EU investment through the Framework Programmes (FP7, Horizon 2020) per country of beneficiary

Nr of Horizon Nr of Number of Nr of 2020 % of total FP7 Horizon Nr of FP7 Horizon Horizon Horizon Horizon % of participations Horizon % of Horizon investm 2020 participatio % of 2020 2020 2020 % of European 2020 total in signed 2020 signed total 2020 ent per participatio ns in signed total projects investmen investment total Innovation Country participatio Horizon grants grants with Horizon investme year, EU15/EU13 ns in signed grants per FP7 with at t, signed per year, FP7 Scoreboard code ns in signed 2020 Thousands at least one 2020 nt per signed Grouping grants per country of particip least 1 grants signed invest- Country Group grants per participat Scientists participant invest EUR of grants country of beneficiary ations participant (EUR grants (EUR ment 2018 country of ions and Engineers from ment GERD (EUR beneficiary per year from million) million) beneficiary in the country million) per year country population AT 2460 547 2.8% 519 2.6% 8.7 1606 8.3% 944 210 2.9% 1.9% 170 2.6% STRONG EU15 BE 3901 867 4.4% 857 4.3% 8.0 2467 12.8% 1574 350 4.7% 3.3% 277 4.3% STRONG EU15 BG 410 91 0.5% 102 0.5% 2.1 300 1.6% 63 14 0.2% 3.7% 14 0.2% MODEST EU13 CY 428 95 0.5% 67 0.3% 15.5 326 1.7% 128 28 0.4% 31.1% 13 0.2% MODERATE EU13 CZ 898 200 1.0% 207 1.0% 2.7 678 3.5% 206 46 0.6% 1.5% 41 0.6% MODERATE EU13 DE 10635 2363 12.0% 2668 13.5% 3.3 5299 27.5% 5282 1174 15.9% 1.3% 1022 15.7% STRONG EU15 DK 1995 443 2.3% 409 2.1% 7.0 1457 7.6% 842 187 2.5% 2.3% 154 2.4% LEADER EU15 EE 413 92 0.5% 80 0.4% 8.5 327 1.7% 123 27 0.4% 10.1% 14 0.2% MODERATE EU13 EL 2491 554 2.8% 541 2.7% 9.7 1468 7.6% 751 167 2.3% 9.5% 146 2.2% MODERATE EU15 ES 9154 2034 10.4% 1669 8.4% 6.6 4654 24.1% 2981 662 9.0% 5.0% 471 7.3% MODERATE EU15 FI 1742 387 2.0% 417 2.1% 6.3 1157 6.0% 735 163 2.2% 2.8% 125 1.9% LEADER EU15 FR 8532 1896 9.7% 1867 9.4% 4.8 4211 21.8% 3493 776 10.5% 1.5% 744 11.5% STRONG EU15 HR 385 86 0.4% 59 0.3% 4.3 288 1.5% 59 13 0.2% 3.3% 13 0.2% MODERATE EU13 HU 762 169 0.9% 237 1.2% 3.1 591 3.1% 208 46 0.6% 3.4% 41 0.6% MODERATE EU13 IE 1441 320 1.6% 287 1.5% 6.2 1087 5.6% 565 126 1.7% 3.9% 90 1.4% STRONG EU15 IT 8419 1871 9.5% 1748 8.8% 7.9 4144 21.5% 2735 608 8.3% 2.8% 519 8.0% MODERATE EU15 LT 293 65 0.3% 61 0.3% 2.9 240 1.2% 42 9 0.1% 2.9% 7 0.1% MODERATE EU13 LU 270 60 0.3% 37 0.2% 10.6 233 1.2% 86 19 0.3% 2.9% 9 0.1% LEADER EU15 LV 255 57 0.3% 48 0.2% 4.9 213 1.1% 49 11 0.1% 9.9% 7 0.1% MODERATE EU13 MT 126 28 0.1% 29 0.1% 9.8 99 0.5% 20 5 0.1% 7.4% 3 0.0% MODERATE EU13 NL 5437 1208 6.2% 1213 6.1% 6.6 3202 16.6% 2485 552 7.5% 3.9% 489 7.5% LEADER EU15 PL 1374 305 1.6% 319 1.6% 1.1 981 5.1% 306 68 0.9% 1.7% 63 1.0% MODERATE EU13 PT 1928 428 2.2% 349 1.8% 5.4 1230 6.4% 537 119 1.6% 5.1% 75 1.2% MODERATE EU15 RO 759 169 0.9% 156 0.8% 1.4 516 2.7% 114 25 0.3% 3.1% 19 0.3% MODEST EU13 SE 2682 596 3.0% 659 3.3% 4.6 1782 9.2% 1162 258 3.5% 1.7% 250 3.9% LEADER EU15 SI 733 163 0.8% 134 0.7% 9.6 520 2.7% 193 43 0.6% 5.3% 24 0.4% STRONG EU13 SK 370 82 0.4% 71 0.4% 3.7 275 1.4% 84 19 0.3% 2.9% 11 0.2% MODERATE EU13 UK 10336 2297 11.7% 2603 13.2% 3.0 6376 33.0% 4729 1051 14.3% 2.6% 1003 15.4% LEADER EU15

EU13 7206 1601 8.2% 1569 7.9% n.a 17319 17% 1595 354 4.8% 2.9% 272 4.2%

EU15 71423 15872 80.8% 15843 80% n.a 3209 90% 28900 6422 87% 2% 5543 85%

EU28 78629 17473 89.0% 17412 89% n.a n.a n.a 30495 6777 92% 2% 5815 90%

Associated 6336 1408 7.2% 1629 8% n.a 3816 20% 2467 37 1% n.a 584 9% Countries 25

Total 88374 19639 100% 19758 100% n.a 19292 n.a 33130 7362 100% n.a 6486 100%

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, Cut-off date 1 July 2018, Eurostat data for GERD and HRST, European Innovation Scoreboard 2018

Table 3 EU investment through Horizon 2020 by type of action (EUR million)

European European Marie Research and Coordination COFUND- Pre- Public Innovation EU15/ Innovation SME ERA-NET- Research Skłodowska- Innovation and Support European Joint Commercial Procurement of TOTAL Scoreboard EU13 Actions Instrument Cofund Council Curie Actions Actions Actions Programming Procurement Innovation Country Group Grouping 2018 DK 156 157 255 145 33 74 7 13 2 0 842 LEADER EU15 FI 145 48 255 160 34 75 6 10 4 0 735 LEADER EU15 LU 15 6 38 11 13 1 0 2 0 0 86 LEADER EU15 NL 617 268 925 435 116 77 10 29 7 1 2485 LEADER EU15 SE 239 115 452 196 41 82 8 26 5 0 1162 LEADER EU15 UK 1408 692 1666 589 173 136 36 24 5 1 4729 LEADER EU15 AT 203 78 348 199 66 30 3 17 0 0 945 STRONG EU15 BE 235 138 535 336 276 14 12 15 14 0 1574 STRONG EU15 DE 1154 391 1985 1016 220 104 336 66 1 9 5283 STRONG EU15 FR 811 312 1388 634 157 103 38 36 6 9 3493 STRONG EU15 IE 77 101 184 105 26 64 2 8 0 0 565 STRONG EU15 SI 6 12 74 48 28 17 3 4 0 0 193 STRONG EU13 CY 4 12 37 25 46 1 1 2 0 0 128 MODERATE EU13 CZ 32 23 80 29 29 4 6 2 0 0 206 MODERATE EU13 EE 4 7 23 31 32 22 2 2 1 0 123 MODERATE EU13 EL 24 53 401 198 43 12 3 2 17 0 751 MODERATE EU15 ES 387 316 1130 712 128 252 15 30 9 2 2981 MODERATE EU15 HR 3 5 22 11 15 2 0 1 0 0 59 MODERATE EU13 HU 47 12 64 20 35 26 3 2 1 0 208 MODERATE EU13 IT 321 242 1162 646 149 131 44 24 7 10 2735 MODERATE EU15 LT 3 5 13 8 9 5 1 1 0 0 42 MODERATE EU13 LV 0 4 12 6 22 2 1 3 0 0 49 MODERATE EU13 MT 2 1 5 4 6 2 0 0 0 0 20 MODERATE EU13 PL 15 41 118 58 39 22 3 10 0 0 306 MODERATE EU13 PT 84 56 177 117 66 21 5 6 5 0 536 MODERATE EU15 SK 0 6 20 27 26 2 1 2 0 0 84 MODERATE EU13 BG 0 5 16 13 25 2 2 1 0 0 63 MODEST EU13 RO 5 8 43 29 18 0 2 8 0 0 114 MODEST EU13 EU28 5996 3113 11427 5803 1870 1279 546 346 83 31 30495 EU15 5876 2972 10901 5497 1540 1172 523 308 80 31 28900 EU13 120 142 527 307 330 107 23 37 3 0 1595 TOTAL 6924 3347 12225 6168 1979 1406 556 401 94 31 33130 Source DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, Cut-off date 1 July 2018, European Innovation Scoreboard 2018

Table 4 Number of participations in Horizon 2020 by programme part, per country of beneficiary

LEIT- LEIT- Innovatio FTI ERC FET MSCA RI LEIT LEIT-ICT ARF SC1 SC2 SC3 SC4 SC5 SC6 SC7 SEWP SWAFS Euratom TOTAL NMBP SPACE n in SMEs AT 9 140 67 375 66 0 356 156 38 0 48 114 124 259 310 150 81 72 24 59 12 2460 BE 17 170 69 570 104 1 395 277 99 11 60 282 350 353 477 274 121 122 17 50 82 3901 BG 12 1 10 46 28 1 21 12 5 2 23 16 33 77 21 20 14 28 14 16 10 410 CY 1 9 5 64 16 0 44 21 7 0 11 19 13 43 23 31 15 22 63 18 2 428 CZ 3 23 14 114 80 0 105 43 20 1 20 41 48 97 115 39 25 14 24 18 54 898 DE 62 781 342 1852 443 0 1550 784 241 3 212 673 514 854 1133 475 186 233 90 96 111 10635 DK 4 107 43 545 63 0 159 93 21 0 47 169 177 221 112 114 52 27 11 25 5 1995 EE 0 4 6 46 23 1 41 17 8 0 19 26 38 52 15 24 31 22 23 13 4 413 EL 15 28 63 304 137 0 463 169 54 1 50 137 161 214 186 154 114 167 8 52 14 2491 ES 58 314 221 1473 274 2 1320 780 227 3 280 530 672 921 823 601 155 282 24 97 97 9154 FI 4 93 39 239 100 0 299 107 21 2 30 105 127 149 100 140 48 56 14 25 44 1742 FR 36 679 309 1468 434 1 1089 491 282 4 145 515 619 507 920 382 121 205 36 54 234 8532 HR 2 3 4 27 20 0 16 3 2 0 22 31 40 74 44 37 16 8 15 11 10 385 HU 1 32 14 91 47 1 77 24 6 0 34 40 91 53 75 49 43 20 23 18 23 762 IE 11 58 14 344 66 0 206 96 13 1 30 85 140 103 73 72 38 57 8 22 4 1441 IT 54 299 243 1253 358 2 1049 696 246 4 215 492 639 785 880 530 200 290 43 74 67 8419 LT 1 1 2 36 12 1 24 12 7 0 16 6 22 42 38 12 22 4 9 12 14 293 LU 1 8 1 29 8 0 42 23 3 0 14 20 2 22 29 7 15 25 13 8 0 270 LV 0 0 3 20 14 0 20 14 5 2 7 20 34 53 12 7 12 9 12 5 6 255 MT 0 1 1 19 9 0 6 5 2 2 11 2 7 12 12 9 5 5 7 11 0 126 NL 55 436 117 1134 295 1 592 282 71 6 47 473 420 389 460 320 98 118 27 60 36 5437 PL 2 12 25 202 100 0 139 83 27 0 99 65 98 116 101 88 54 69 31 31 32 1374 PT 7 63 26 319 94 1 235 129 38 0 50 90 163 188 106 164 50 95 69 32 9 1928 RO 4 6 5 69 36 1 65 26 12 0 71 35 67 101 58 76 19 62 16 9 21 759 SE 11 161 76 491 130 0 282 158 33 0 56 203 170 230 319 179 56 44 20 18 43 2682 SI 11 7 13 69 34 0 80 50 10 0 31 48 53 95 63 65 22 19 29 17 17 733 SK 2 1 5 43 25 1 33 16 4 0 12 16 30 41 45 24 23 10 21 7 11 370 UK 92 999 323 3082 406 1 974 490 191 3 107 748 489 595 674 435 223 265 61 84 93 10336 EU13 39 100 107 846 444 6 671 326 115 7 376 365 574 856 622 481 301 292 287 186 204 7206 EU15 436 4336 1953 13478 2978 9 9011 4731 1578 38 1391 4636 4767 5790 6602 3997 1558 2058 465 756 851 71423 EU28 475 4436 2060 14324 3422 15 9682 5057 1693 45 1767 5001 5341 6646 7224 4478 1859 2350 752 942 1055 78629 Associated 39 620 184 1177 328 2 682 369 98 2 193 386 484 524 380 359 158 187 30 90 44 6336 Countries Third Countries 0 39 10 1852 128 0 179 64 52 0 5 226 292 47 62 236 145 14 2 36 19 3409 Total 514 5095 2254 17353 3878 17 10543 5490 1843 47 1965 5613 6117 7217 7666 5073 2162 2551 784 1068 1118 88374

Table 5 EU investment through Horizon 2020 by programme part, per country of beneficiary (EUR million)

Innovat LEIT- LEIT- LEIT- FTI ERC FET MSCA RI LEIT ARF ion in SC1 SC2 SC3 SC4 SC5 SC6 SC7 SEWP SWAFS Euratom TOTAL ICT NMBP SPACE SMEs AT 4.6 203.2 30.6 79.0 19.2 0.0 126.2 70.7 10.8 0.0 7.2 52.2 30.6 97.3 102.5 41.3 21.7 25.3 4.8 15.3 2.1 944 BE 4.9 234.8 29.4 137.8 28.7 0.2 207.1 104.2 26.1 2.9 9.3 102.9 87.1 130.6 131.3 78.1 32.7 40.7 139.8* 15.3 29.4 1574 BG 0.1 0.2 1.5 5.0 2.4 0.0 5.4 2.3 0.3 0.1 0.3 2.3 4.7 10.5 2.3 3.5 1.6 3.4 14.2 2.5 0.7 63 CY 0.0 3.8 3.2 12.9 2.7 0.0 12.7 5.8 1.2 0.0 0.5 5.5 3.3 8.2 9.6 5.8 3.9 5.2 39.5 3.0 0.1 128 CZ 0.9 32.1 4.6 23.1 12.7 0.0 22.1 10.1 2.4 0.0 0.9 8.0 9.0 17.5 20.4 7.0 4.3 2.6 14.9 3.0 9.9 205 DE 24.8 1154.1 185.0 391.1 191.7 0.0 681.9 360.2 86.2 0.8 25.9 335.9 198.3 432.4 494.6 192.6 59.0 81.3 22.2 25.2 339.3 5282 DK 1.6 156.2 22.8 157.4 20.5 0.0 64.3 43.9 4.2 0.0 9.5 78.8 61.4 98.1 52.7 37.2 14.7 8.9 2.0 6.5 1.1 842 EE 0.0 3.6 1.3 7.4 2.1 0.3 10.8 8.5 1.0 0.0 1.7 9.4 15.3 16.4 2.8 5.3 5.8 4.5 24.0 2.2 1.0 123 EL 5.2 23.7 29.8 54.3 40.1 0.0 165.9 66.1 10.3 0.1 4.0 50.3 40.5 56.3 55.1 44.2 23.8 69.8 1.1 8.4 2.0 751 ES 22.5 387.0 95.3 318.2 62.3 0.4 387.6 289.4 68.6 0.1 32.3 218.1 182.4 316.7 247.6 190.0 33.8 83.8 3.9 20.5 20.7 2981 FI 2.2 145.2 16.0 48.2 30.8 0.0 113.7 52.3 6.2 0.1 10.7 42.8 45.3 76.4 41.6 51.6 13.8 19.6 2.2 4.5 12.0 735 FR 13.8 812.1 124.2 312.7 155.4 0.2 439.4 187.5 97.8 0.3 23.4 236.4 173.0 254.5 354.1 113.3 28.5 77.2 4.1 10.1 75.4 3493 HR 0.7 2.6 1.6 4.9 2.3 0.0 2.9 0.8 0.2 0.0 1.6 7.2 5.7 9.0 4.6 5.0 2.2 0.4 5.4 1.4 0.4 59 HU 0.4 47.2 5.6 12.4 7.6 0.2 21.8 9.5 0.9 0.0 2.6 12.5 14.8 10.6 13.2 9.4 6.7 2.9 22.9 2.6 3.9 207 IE 8.5 76.6 6.8 101.4 16.6 0.0 78.9 46.7 5.2 0.0 6.7 52.4 63.9 39.1 15.1 16.9 8.9 14.2 1.5 5.5 0.4 565 IT 20.8 321.1 99.9 246.1 115.4 0.3 325.7 238.6 72.7 1.5 16.9 204.9 189.1 254.9 268.2 152.1 49.2 85.8 7.0 17.8 46.7 2735 LT 0.4 2.5 0.2 5.0 0.9 0.1 4.8 2.9 1.6 0.0 0.5 0.3 2.3 5.0 3.6 2.1 4.1 0.3 3.3 1.1 1.4 42 LU 0.3 15.2 0.2 5.7 1.5 0.0 11.2 8.2 0.8 0.0 3.3 7.1 0.5 3.5 9.0 1.5 3.5 8.6 4.4 1.1 0.0 86 LV 0.0 0.0 0.8 3.9 1.1 0.0 4.0 2.8 0.3 0.4 0.3 2.6 4.4 7.2 4.6 0.8 1.6 0.6 13.0 0.2 0.5 49 MT 0.0 1.6 0.6 1.9 0.3 0.0 0.7 0.7 0.2 0.8 0.1 0.4 1.1 1.1 2.7 1.6 0.8 2.8 2.0 0.8 0.0 20 NL 23.8 617.2 56.7 268.9 101.2 0.1 272.5 121.7 17.4 0.9 17.7 285.9 159.5 161.1 163.6 117.3 30.0 37.9 5.7 16.0 9.8 2485 PL 0.4 14.8 9.1 43.5 20.4 0.0 47.8 23.4 4.1 0.0 8.7 19.9 17.8 21.0 12.8 17.9 8.3 13.6 13.4 4.9 4.5 306 PT 2.4 84.2 9.4 57.1 13.4 0.1 58.6 42.6 13.3 0.0 4.7 21.9 35.1 55.3 24.0 42.2 8.3 23.5 32.9 4.1 3.5 536 RO 0.7 5.1 1.2 8.3 4.2 0.0 12.6 4.7 1.0 0.0 1.3 7.1 9.5 16.8 8.2 9.8 2.6 12.1 6.0 1.1 1.9 114 SE 5.3 239.4 57.2 114.7 45.2 0.0 109.3 67.0 10.1 0.0 7.6 110.3 58.2 103.0 107.6 70.3 14.8 18.1 6.9 3.7 12.3 1162 SI 4.9 6.4 3.2 12.5 4.1 0.0 18.6 21.9 2.7 0.0 0.5 10.3 11.8 27.5 19.7 16.8 3.3 4.5 18.2 2.2 3.5 193 SK 0.4 0.4 1.2 6.5 2.7 0.1 6.1 3.7 0.4 0.0 0.5 1.9 25.0 4.7 4.3 2.7 2.8 1.4 17.1 0.6 1.6 84 UK 40.6 1410.0 175.9 694.0 191.9 0.1 398.7 206.0 54.3 0.5 12.7 420.6 157.1 293.3 250.5 172.0 77.6 84.5 23.0 19.1 45.7 4729 EU13 9 120 34 147 63 1 170 97 16 1 19 88 125 155 109 88 48 54 194 25 29 1595 EU15 181 5880 939 2986 1034 1 3441 1905 484 7 192 2221 1482 2373 2318 1321 420 679 261 173 600 28900 EU28 190 6000 973 3134 1097 2 3611 2002 500 8 211 2308 1607 2528 2426 1408 468 734 455 198 630 30495 AC countries 0 9 920 125 236 75 0 204 121 17 0 0 106 146 164 93 95 30 56 19 12 2467 Third Countries 0 0 8 1 2 14 0 12 3 4 0 0 49 21 6 2 26 14 1 0 3 167 Total 191 6009 1902 3260 1335 92 3611 2218 624 30 211 2308 1762 2695 2596 1504 589 777 512 217 645 33130 *Includes payments from SEWP to the COST Association located in Belgium Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, Cut-off date 1 July 2018

Table 6 Number of participations and EU investment (EUR million) through Horizon 2020 for specific types of partnerships, per country of beneficiary per year

Future and Emerging COFUND-European Joint Contractual Public-Private Joint Undertakings* ERA-NET-Cofund TOTAL per year Technologies-Flagships** Programming Partnerships*** European Innovation EU EU EU EU EU Total Total EU Participation Participation Participation Participation Participation EU28 Scoreboard Country Investment Investment Investment Investment Investment Participation Investment Group 2018 AT 62 20 1 1 12 4 3 1 19 8 97 34 EU15 STRONG BE 72 29 2 0 22 3 3 3 34 13 133 49 EU15 STRONG BG 2 0 0 0 2 0 2 0 1 0 7 1 EU13 MODEST CY 0 0 0 0 3 0 1 0 5 1 9 2 EU13 MODERATE CZ 24 4 0 0 3 0 4 1 6 2 38 7 EU13 MODERATE DE 196 92 6 4 25 15 5 75 127 57 359 243 EU15 STRONG DK 26 7 1 0 7 3 3 2 8 4 44 15 EU15 LEADER EE 2 3 0 0 5 0 1 0 2 0 10 4 EU13 MODERATE EL 14 4 1 0 3 0 3 1 28 12 50 17 EU15 MODERATE ES 139 41 3 1 24 7 8 3 108 35 282 87 EU15 MODERATE FI 39 12 1 0 9 2 3 1 16 8 67 24 EU15 LEADER FR 166 74 4 2 24 8 14 8 69 25 277 118 EU15 STRONG HR 7 1 0 0 1 0 2 0 1 0 12 1 EU13 MODERATE HU 10 1 1 0 3 0 2 1 4 1 20 3 EU13 MODERATE IE 20 7 0 0 8 2 1 0 9 3 38 12 EU15 STRONG IT 136 55 3 1 19 5 8 10 96 34 262 105 EU15 MODERATE LT 7 0 0 0 3 0 1 0 2 0 14 1 EU13 MODERATE LU 2 0 0 0 2 0 0 0 2 1 6 2 EU15 LEADER LV 3 1 0 0 6 1 1 0 2 0 12 2 EU13 MODERATE MT 1 0 0 0 0 0 0 0 0 0 2 0 EU13 MODERATE NL 103 47 2 1 16 7 4 2 34 13 159 70 EU15 LEADER PL 14 3 0 0 9 2 3 1 14 4 41 10 EU13 MODERATE PT 18 3 1 0 8 1 2 1 14 5 44 11 EU15 MODERATE RO 6 0 0 0 9 2 3 1 5 1 23 4 EU13 MODEST SE 43 12 2 1 11 6 3 2 24 10 83 30 EU15 LEADER SI 5 1 0 0 5 1 2 1 8 3 21 5 EU13 STRONG SK 13 5 0 0 4 0 1 0 2 0 20 6 EU13 MODERATE UK 90 45 4 2 8 5 3 8 71 33 177 92 EU15 LEADER EU13 96 20 2 1 52 8 24 5 53 14 227 48 EU15 1126 449 28 14 198 69 64 116 662 261 2078 908 EU28 1221 469 31 15 250 77 88 121 715 274 2306 956 Non EU 93 24 5 7 8 12 4 2 50 15 160 60 TOTAL 1314 493 36 21 36 89 92 123 765 290 2244 1016 Notes: *JU: Ecsel, BBI, FCH, JTI, SESAR, CS2 ; ** FET-Flagship 2014,2016,2017 & HBP ; ***cPPP: FoF, EeB, SPIRE, FET HPC. Includes Euratom, excludes Art. 185 and KICs as those actions are not included in CORDA. Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, Cut-off date 1 July 2018, European Innovation Scoreboard 2018

Table 7 Share of Horizon 2020 projects with at least one participant from each country group per programme part (incl. mono-beneficiary projects)

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, Cut-off date 1 July 2018

Table 8 EU investment (EUR million) in “Widening” and “Non-Widening” countries through specific actions under Horizon 2020 under the Programme part ‘Spreading Excellence and Widening Participation’

Budget Distribution Widening Non-Widening ERA Chairs 2017 32 0

Twinning 2017 14 16 ©European Union, 2020 Teaming Phase 1 2017 7 5 Teaming Phase 2 111 29 Teaming Phase 1 2014 0 6 Twinning 2015 32 35 ERA Chairs 2014 34 0 TOTAL* 238 90 % budget distribution 73% 27%

Notes: The Member States currently eligible under Horizon 2020 for Widening support are: Bulgaria, Croatia, Cyprus, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Luxembourg, Malta, Poland, Portugal, Romania, Slovakia and Slovenia. The Associated Countries eligible for Widening support are (subject to valid association agreements of third countries with Horizon 2020): Albania, Armenia, Bosnia and Herzegovina, Faroe Islands, Former Yugoslav Republic of Macedonia, Georgia, Moldova, Montenegro, Serbia, Tunisia, Turkey and Ukraine. *Excludes payments from SEWP to the COST Association located in Belgium Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit

Other monitoring flash reports available here: https://ec.europa.eu/info/publications/horizon-2020-monitoring-flash_en

#2 Dynamic Network Analysis #3 International Cooperation #4 Patents in FP #5 Sustainable Development Goals

2 DNAC NETWOR ANALSS

by Julien Ravet

FROM HORIZON 2020 TO HORIZON EUROPE MONITORING FLASH DYNAMIC NETWORK ANALYSIS

November 2018 FROM FP6 TO HORIZON 2020 Key overview data 5 million 5/1 7,509 23,664 149 collaborations within ratio of new to collaborative participants countries represented FP6, FP7 and Horizon 2020 maintained projects launched in collaborated under in collaborative projects projects collaborations in the first the first four years of the first four years of under Horizon 2020 four years of Horizon 2020 Horizon 2020 Horizon 2020 The size of the Horizon 2020 collaboration network is massive: since 2014, Horizon 2020 has funded more than 7,500 collaborative projects among 23,664 participants from 149 countries, which results in almost 1.5 million of one-to-one collaborations. For every maintained collaboration, there are almost 5 new ones. Since FP6, 5 million collaborations have been generated by the Framework Programmes.

A MASSIVE NETWORK FOR R&I ACTIVITIES Key overall messages

. The most central countries in the network are also the largest ones: Germany, France, the UK, Italy, and Spain. . When taking into account the size of countries, some “punch above their weight”: the most central country in Horizon 2020 is Finland, followed by Slovenia. While some new Member States like Cyprus, Estonia, Malta and Slovenia are as central as EU-15 countries, other EU-13 countries remain at the bottom of the connectivity rankings.. . Slovenia, Luxembourg, Croatia, Portugal and Cyprus show the most striking increases in terms of size- normalised centrality between FP7 and Horizon 2020, while the position of the UK and Hungary dropped. . Although EU-15 participants have reduced their collaborations with EU-13 participants in FP7, this trend appears to be reverting in Horizon 2020. . Geographical and cultural proximities between participants seem to play an important role in shaping the structure of the Horizon 2020 collaboration network.

Introduction

A key EU Added Value of the EU Framework Programmes for Research and Innovation (R&I) consists in the creation of transnational and multidisciplinary networks (European Commission, 2017 and 2018). The Framework Programmes offer unique collaboration and networking opportunities between researchers. Collaborations within the network generate spillover effects and knowledge sharing while bringing the R&I effort in Europe closer to the critical mass required to tackle global societal challenges. The majority of the Horizon 2020 budget is spent on supporting such collaboration through collaborative R&I projects. However, to fully reap the benefits of collaborative R&I, it is important that the network remains open and easily accessible to new participants. In this context, a good understanding of the way researchers collaborate within the Programme is crucial.

The Horizon 2020 Interim Evaluation already provides insights into the collaborations between researchers based on publications1 and project data. In particular, it suggests that collaboration patterns may have evolved between the 7th Framework Programme (FP7) and Horizon 2020. A previous study also examined the evolution of collaborations between the 6th Framework Programme (FP6) and FP7 (Science Metrix, 2015; with 40% of the projects completed in FP7).

This monitoring flash further explores certain aspects of the collaborations between participants and provides additional evidence related the dynamic evolution of the network of participants to the Programme. While the complexity of such a large network can be examined from different angles, this flash focuses on cross-country collaborations. In particular, the analysis highlights how the situation of entities in participating countries has changed over the last decade.

The analysis is based on monitoring data of Horizon 2020 and its predecessor programmes, FP6 and FP7, covering the 2003-2017 period2. The data covers collaborative projects3 launched during the first four years of implementation of Horizon 2020, and the full implementation of FP6 and FP74. The data is stored in the Common Research Data Warehouse (CORDA), an internal database maintained by DG Research and Innovation. For this paper, country groups (i.e. EU-15, EU-13, associated countries and third countries) are based on the situation in Horizon 2020. This flash is based on the main results from a more extended study, Balland and Ravet (2018)5.

1 Elsevier (2017). 2 Year of signature of the contract. Cut-off date for Horizon 2020 is 1/1/2018. 3 Data include all evaluated calls for collaborative projects. Projects under Public-Public Partnerships, EIT’s Knowledge and Innovation Communities (KICs) and direct actions of the Joint Research Centre are not included. 4 Projects with incomplete data on signature date, duration and participant identifier were removed from the analysis (the final dataset includes about 99.1% of the initial dataset of collaborative projects). 5 Balland, P.A., and Ravet, J. (2018). Dynamic Network Analysis of the EU R&I Framework Programme. European Commission report, forthcoming. More details, including on the technical aspects, are available in the study. Calculations were produced by Pierre-Alexandre Balland (Utrecht University and Massachusetts Institute of Technology).

Overview of the Horizon 2020 network

Figure 1 Structure of the H2020 Collaboration Network

Note: This graph represents the backbone of H2020. Nodes are countries, and links represent strong6 connections based on Horizon 2020 projects. EU- 15 countries are represented in blue, EU-13 countries are represented in orange and Associated Countries (AC) are represented in green. The size of the nodes expresses their centrality, and the width of the links expresses the number of collaborations. Source: CORDA data. Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit

The size of the Horizon 2020 collaboration network is massive. Since 2014, Horizon 2020 has been funding a very large number of collaborative projects, which involved a very large network of collaborations between R&I stakeholders. Over 2014-2017, Horizon 2020 funded more than 7,500 collaborative projects among 23,664 participants from 149 countries, which results in almost 1.5 million of one-to-one opportunities to collaborate7.

The strongest connections that emerged out of Horizon 2020 are represented as a country-country graph in Figure 18. The figure shows two types of connections: (i) the single strongest connection of each country to another country, and (ii) the top 40 strongest connections in the network. Centrality can be defined as the importance of a country in the network. This importance as such can have different meanings, hence different definitions, with the most straightforward definition being based on the number of connections of a country’s participants in the whole network. The size of the nodes is proportional to the centrality of the country. The core of the network is mainly composed of EU-15 participants. Germany, France, the UK, Italy, and Spain appear to be key players in the network of participations to Horizon 2020.

6 The links displayed on this graph with N actors combines the N-1 links of a maximum spanning tree (MST) and the N-1 strongest links of the overall network. The MST represents the backbone of a weighted network and is based on three rules. First, it keeps only N-1 links from a network with N actors. Second, rule #1 should be satisfied while keeping the strongest links. Third, rule #1 and #2 should be satisfied without creating any isolate in the network. 7 Before Horizon 2020, FP6 and FP7 funded respectively 5,912 and 12,493 collaborative projects, which correspond to 1,305,305 and 1,989,450 collaborations between participants. 8 Third countries are excluded from this visualisation. They are analysed in Balland and Ravet (2018). The analysis is based on the network of participations without any threshold in the number of connections between two participants. Balland and Ravet (2018) use two sets of data for robustness: one without threshold and one based on connections in at least two projects in order to exclude one-off collaborations. The key messages are similar between the datasets. This Flash is based on data that include all connections.

EU-13 participants have a substantial number of collaborations with the largest players in the network, which are participants from EU-15 countries9. As a result, German participants are frequent partners of several EU-13 countries, such as Czech Republic, Hungary, Latvia, Lithuania and Slovakia. Croatia, Malta and Romania present strong ties with Italy, while Bulgaria, Cyprus, Estonia and Slovenia tend to connect with Spanish participants. Important collaborators of Polish participants are French participants.

Evolution of general characteristics of the network

The nature of the network has changed as follows since FP6: . The network of participations to the Framework Programmes seems to be very dynamic over time10. Between FP6 and FP7, about new 1,226,970 connections between partners were created, while 166,508 connections were maintained and 772,822 were lost. Between FP7 and the first four years of Horizon 2020, 909,444 new connections were made, against 195,474 maintained and 1,198,004 lost.

. The network is more dynamic for EU-13 countries than for EU-15 countries. EU-13 countries participants seem to have a higher propensity to be involved in new collaborations than participants from EU-15 countries, which is especially true in Horizon 2020.

. The network tends to be opening to less connected participants since FP6. On average, participants are slightly less central11 in the network in FP7 and Horizon 2020 compared to FP6. This might signal the entry of smaller players.

. Participant acting as hubs (i.e. with high centrality) also seem to connect more likely with other types of participants (non-hubs, with low centrality)12. This suggests that key actors in the network have maintained a certain level of openness to other participants throughout the different programmes.

. Network inequality coefficients13 are particularly stable over time. This suggests that a few organisations have many connections, while most organisations have only a few - this is a general tendency of real-world complex networks. This aspect of the network has not been reinforced over time.

. The average path length between participants has remained close to 3, meaning that on average a participant can be connected to any other participant in the network within 3 connections (“degrees of separation”). This measure is relatively small, indicating a highly-connected network in general. The average path length has not changed much over time.

9 Top 5 EU-15 participants that present the largest numbers of collaborations with EU-13 participants are Fraunhofer (DE), CNR (IT), CNRS (FR), CEA (FR) and VTT (FI). Top 5 EU-15 participants that present the highest share of collaborations with EU-13 participants in their collaborations are ENEA (IT), NERC (UK), CINECA (IT), UoA (EL) and JUELICH (DE). 10 Based on Jaccard coefficients (Ripley et al., 2016). 11 Centrality can be defined as the importance of a node (here a participant) in the network. This importance as such can have different meanings, hence different definitions. Measures used here are degree centrality and eigenvector centrality. 12 Based on assortativity coefficients (extent to which participants in the network associate with other participants with similar degree centrality). The Framework Programmes are disassortative networks in the sense that central nodes tend to connect to less central nodes. 13 Network Gini coefficient of the degree distribution.

Evolution of the position of countries in the network Which countries are becoming more central? This section presents network indicators computed at the participant level, and averaged or aggregated at the country level. Figure 2 details the evolution of country rankings based on centrality measures (here eigenvector centrality coefficients14). Germany is both the largest participant in the Framework Programme and the most central country in the network. After Germany, France and Italy are the most central countries in Horizon 2020. While the UK was more central than France and Italy in FP6 and FP7, its central position worsened in Horizon 2020. Greece, Portugal and Ireland have improved their centrality in the network between FP7 and Horizon 2020 according to this ranking. The chart also confirms that participants from EU-15 countries tend to be more central than their EU-13 counterparts: the bottom of the chart is occupied by a majority of EU- 13 countries. Only Croatia seems to have significantly improved its position since FP6.

Figure 2 Network positions of participants by EU country (ranking)

(total) Eigenvector centrality rankings

Framework Programme

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, Corda data.

However, these measures are absolute and are significantly influenced by country size15. Figure 1 and 2 illustrate this size effect in the network: the most connected countries are also the largest ones. This is also shown in Figure 3 (based on million inhabitants). The most connected country is Germany, with around 12% of the collaborations within the network involving German participants, followed by Spain (11%), Italy (10%),

14 This indicator measures the influence of a country in the network by examining whether participants of a country are linked to other important participants (i.e. participants with many connections). 15 To ensure robustness, other variables describing country size have been tested, such as the national population of researchers (source: Eurostat). This does not affect the key messages from the analysis. However, using population reduces data noise over time and ensures reliability in the evolution of the ranking.

and France (10%). Overall, 79.3% of the collaborations involve participants from EU-15 countries against 9.8% for EU-13 countries (and respectively 6.6% for associated countries and 4.2% for third countries). Poland is the EU-13 country with most connections (1.8% of all connections).

Figure 3 Country size and share of connections under Horizon 2020

(%)

5 Share of Horizon 2020 connections

0 2 4 6 8

Country size (million inhabitants)

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, CORDA data (Horizon 2020) and World Bank (country population).

Normalisation for size leads to an overall different picture. Figure 4 presents the evolution of the same centrality coefficients as in Figure 2 when normalising by country size (as measured by population). Different trends can be observed. The most central country, relative to its size, is actually Finland. Some EU-13 countries also appear to be very central in the network for their size: Slovenia is now the second most central country in the network after normalisation for size effect. This was not the case in previous programmes: Slovenia was ranked 5th in FP6 and 8th in FP7 in terms of centrality. This is the most striking increase observed within all EU countries. Luxembourg, the Netherlands, Belgium, Sweden and Denmark are next in terms of size-normalised centrality measure. Among EU-13 countries, Cyprus and Estonia also present strong centrality after normalisation.

Hence EU-15 and EU-13 groups are not homogenous groups, with some EU-13 countries being more central, relative to their size, than most EU-15 countries. The position of the UK and Hungary in this ranking dropped significantly between FP7 and Horizon 202016. Still, several EU-13 countries are consistently found at the bottom of the ranking over the period.

16 The position of Malta also decreased significantly over the same period, but it follows a significant increase in FP7 and the position of small countries is more volatile in the ranking.

Figure 4 Network positions of participants by EU country normalised by country size

) (normalised by country size Eigenvector centrality rankings

Framework Programme

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, CORDA data (Framework Programme) and World Bank (country population).

Opening of EU-15 to EU-13 participants In terms of dynamic, while EU-15 participants seem to have reduced their collaborations with EU-13 participants between FP6 and FP7, they appear to have opened up to EU-13 participants in Horizon 2020. In FP6, the percentage of connections between EU-15 participants and EU-13 participants was 15.3% of all collaborations from EU-15 participants (Figure 5). While this percentage decreased to 13.1% during FP7, it increased again to 14.5% in Horizon 2020. Hence, while the opening of EU-15 countries to EU-13 countries seems to have worsened during FP7, the situation has improved with Horizon 2020. In parallel, the share of collaborations between EU-13 participants with other EU-13 has been stable since FP6.

Figure 5. Connections with EU-13 participants as a percentage of all connections of EU-15 participants

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit CORDA data.

The evolution of these collaborations between EU-15 and EU-13 countries are detailed for each EU-15 country in Figure 617. While there is a clear general decrease in the collaborations with EU-13 participants between FP6 and FP7, almost all EU-15 countries collaborate more often with EU-13 participants in Horizon 2020 compared to FP7. The only exceptions are Luxembourg and the United Kingdom, which are also respectively the countries with the largest (13.3% in Horizon 2020) and the smallest share of connections (7.5%) with EU-13 participants. Since FP6, this trend has been continuously negative only for the UK and continuously positive only for Greece.

Figure 6 Connections with EU-13 countries as a percentage of all connections

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, CORDA data.

The top EU-15 participants that present the largest numbers of collaborations with EU-13 participants in Horizon 2020 are Fraunhofer (DE), CNR (IT), CNRS (FR), CEA (FR) and VTT (FI). These participants are the most important actors in terms of bridging EU-15 and EU-13 countries. Top 5 EU-15 participants that present the highest share of collaborations with EU-13 participants in their collaborations are ENEA (IT), NERC (UK), CINECA (IT), UoA (EL) and JUELICH (DE).

17 The patterns in Figure 5 and Figure 6 are qualitatively similar. But due to collaborations within country groups, the aggregated values do not numerically correspond to the average of countries.

Preferences in collaborations

Figure 7 shows the country relatedness network, which expresses collaboration preferences between countries. To compute this relatedness, the number of connections between two countries is divided by the number of connections expected by chance18, i.e. based on the amount of participations of both countries (Hidalgo et al. 2007, Balland et al. 2018). In Figure 7, the top four strongest connections of each country are represented. As a result, participants appear to show very specific preferences in their cross-country collaborations. Several clusters of countries can be observed19. Countries in a same cluster of strong preferences are represented by the same colour. Participants from Baltic countries, Czech Republic and Slovakia tend to collaborate more with each other than what would be expected statistically (green cluster). Cyprus, Greece, Ireland, Luxembourg, Malta, and Portugal form another group of preferred connections (yellow cluster). These two groups bridge to some extent the other two clusters, which are formed respectively by large EU-13 countries (pink cluster) and large EU-15 countries (blue cluster). Overall, these preferences show that different forms of proximity, including cultural and geographical proximities tend to shape the structure of the Horizon 2020 network.

Figure 7 Preferred connections – network relatedness (Horizon 2020)

Note: Colours based on community structure (Blondel et al, 2008). The top four strongest connections (after normalisation) of each country are represented. A plain link indicates that the connection is in the top four connections of both countries. A dashed link indicates that the connection is in the top four of one of both countries. Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, CORDA data.

18 Relatedness is computed using the EconGeo software, implemented as a R package (Balland, 2017). 19 Communities within the network are based on the multi-level modularity optimisation algorithm for finding community structure as described by Blondel et al. (2008).

Conclusions

The overall network of participants in Horizon 2020 shows that the most central countries in the network are also the largest ones: Germany, France, the UK, Italy, and Spain. This observation is expected as country size correlates with the number of participations in the Framework Programme and the number of collaborations between participants. Participants from associated countries and third countries are on average less central than EU participants, but these country groups are very heterogeneous. For instance, Switzerland and Norway are very important actors in the network.

When examining the evolution over time and normalising for this size effect, results show a different picture. Some countries punch above their weight: when normalising by country size, the most central country in Horizon 2020 is Finland, followed by Slovenia. Slovenia, Cyprus, Estonia and Malta are as central as EU-15 countries. Still, other EU-13 countries are found at the bottom of the ranking. Slovenia, Luxembourg, Croatia, Portugal and Cyprus show the most striking increases in terms of size-normalised centrality from FP7 to Horizon 2020, while the UK and Hungary dropped positions.

Between FP6 and FP7, EU-15 participants have been reducing to some extent their collaborations to EU-13 participants. However, this trend has reverted in Horizon 2020, as EU-15 countries appear to have opened up to EU-13 participants compared to FP7. This is consistent with the finding of the Monitoring Flash #1, which shows that there are indications that an increasing share of multi-partner Horizon 2020 projects involve at least one EU-13 participant, reversing a downward trend observed under FP7. Moreover, the network of participations to the Framework Programmes appears to be very dynamic over time and tends to be opening to less connected participants. These trends deserve further detailed attention, as well as a more frequent update of observations.

Participants appear to show very specific preferences in their cross-country collaborations. As result, geographical and cultural proximities between participants seem to play an important role in shaping the structure of the Horizon 2020 collaboration network.

Overall, these results show a network that is relatively open, albeit with some persistently peripheral countries. The analysis also presents encouraging trends regarding the openness of the network, in particular between FP7 and Horizon 2020. However, there is still room for improving the connectivity and centrality of several countries, especially countries with lower R&I performance. This calls for continuous emphasis and effort, in particular for these countries, to ensure the openness of the programme’s networks to their entities. This could be achieved through support activities such as organising information/networking campaigns, boosting national capacity building, offering further opportunities to entities for accessing successful R&I projects and established networks, or by supporting matchmaking between potential participants informed by analytics and network affinities.

References Balland, P.A. (2012). Proximity and the Evolution of Collaboration Networks: Evidence from Research and Development Projects within the Global Navigation Satellite System (GNSS) Industry. Regional Studies, 46 (6): 741-756. Balland, P.A. (2017). Economic Geography in R: Introduction to the EconGeo Package. Papers in Evolutionary Economic

Geography, 17 (09): 1-75. © European Union, 2020 Balland, P.A., Boschma, R., Crespo, J. and Rigby, D. (2018). Smart Specialization policy in the EU: Relatedness, Knowledge Complexity and Regional Diversification. Regional Studies, forthcoming. Balland, P.A., and Ravet, J. (2018). Dynamic Network Analysis of the EU R&I Framework Programme. European Commission report. Blondel, V.D., Guillaume, J.-L., Lambiotte, R. and Lefebvre, E. (2008). Fast unfolding of communities in large networks. J. Stat. Mech. P10008. Elsevier (2017). Study on overall output of select geographical group comparators and related FP7- and Horizon 2020- funded publication output. European Commission report. European Commission (2017). Interim Evaluation of Horizon 2020. Staff Working Document. SWD(2017)220. European Commission (2018). Impact Assessment of Horizon Europe. Staff Working Document. SWD(2018)307. Hidalgo, C.A., Klinger, B., Barabási, A.L., Hausmann, R. (2007). The product space conditions the development of nations. Science 317: 482-487. Ripley R., Snijders T.A.B., Boda Z., Voros A., Preciado P. (2016). Manual for RSiena. Available at: http://www.stats.ox.ac.uk/~snijders/siena/RSiena_Manual.pdf. Science-Metrix (2015). Study on Network Analysis of the 7th Framework Programme Participation. European Commission report

Other monitoring flash reports available here: https://ec.europa.eu/info/publications/horizon-2020-monitoring-flash_en

#1 Country Participation #3 International Cooperation #4 Patents in FP #5 Sustainable Development Goals

O OOO

by Julien Ravet, Nelly Bruno, Pantelis Tziveloglou, Michele Ibba

FROM HORIZON 2020 TO HORIZON EUROPE MONITORING FLASH INTERNATIONAL COOPERATION

February 2019

This Monitoring Flash sheds light on the implementation of Horizon 2020 – the European Framework Programme for Research and Innovation 2014-2020 – where international cooperation is promoted and integrated as a cross-cutting priority. The analysis covers the applications and participations from entities located outside the European Union for the first 5 years of Horizon 2020.

HORIZON 2020 Key overview data 124 4,700 18% €300m 7%+4% non-EU countries participants success rate own contribution of participations participating to Horizon from non-EU countries of non-associated of non-associated respectively from 2020 projects in Horizon 2020 countries, higher than countries in Horizon 16 associated + EU average of 15% 2020 projects 108 non-associated countries

A VERY BROAD INTERNATIONAL OUTREACH Key overall messages

. With participants from 124 non-EU countries, Horizon 2020 demonstrates a very broad international outreach attracting talent from around the world in particular from higher education organisations. . There is a certain level of heterogeneity in the cooperation patterns with third countries, reflecting the strategic targeting and diversity of objectives and benefits pursued through international cooperation: - Horizon 2020 is benefitting from excellence worldwide for increasing competitiveness, jointly tackling global challenges and increasing participation in international value chains through the involvement of participants from countries with advanced R&I capabilities. - Horizon 2020 also contributes to the integration of R&I systems in the ERA for countries which have a relative lack of R&I capacity, including through mobility of researchers. . While the international dimension of Horizon 2020 has been reinforced for the second half of Horizon 2020, the identified trends also call for an intensification of international cooperation activities in Horizon Europe.

Introduction

This Monitoring Flash sheds light on the state of play of international cooperation under Horizon 2020 after 5 years of implementation1. This evidence base should inform policy discussions on the Commission proposal for Horizon Europe (2021-2027) and is a natural follow-up of the first two Monitoring Flashes (European Commission, 2018b and 2018c), which focussed on the participations and collaborations of EU Member States2. This Flash provides insights on the participation of non-EU countries, i.e. third countries, to Horizon 2020, based on their status in Horizon 2020: . Associated countries: Albania, Armenia, Bosnia and Herzegovina, Faroe Islands, Georgia, Iceland, Israel, North Macedonia, Moldova, Montenegro, Norway, Serbia, Switzerland, Tunisia, Turkey, Ukraine. . Non-associated third countries: countries that are not EU Member States and not associated to the Framework Programme.

Box 1 How does international cooperation work in Horizon 2020?3

- Association to the programme is limited to countries geographically close to Europe: Enlargement, EFTA and European Neighbourhood Policy countries, as well as countries already associated to FP7. Legal entities from associated countries can participate in actions under the same terms and conditions as entities from Member States.

- Legal entities from non-associated third countries can participate in projects in all parts of the programme, except for mono- beneficiary grants, specific close-to-market innovation activities and access to risk finance.

- Third-country nationals are eligible to apply for European Research Council grants when the host institution is in a Member State or associated country. Third-country nationals are eligible for all Marie Skłodowska-Curie Actions (except for the European Reintegration Panel under the Individual Fellowships scheme).

- Except for a few cases, only participants from low- and middle-income countries are automatically eligible to receive EU funding. EU funding can, exceptionally, be granted to other third-country entities whose participation is deemed essential for carrying out an action.

Why supporting international cooperation under the Framework Programme

It is essential for researchers and innovators in the EU to have access to knowledge, expertise and facilities that lie outside the Union. International collaboration is needed to tackle societal challenges that are global by nature, and it is key to ensure that EU companies stay competitive at the global scale. In this respect, EU-level action can help shaping global multilateral R&I policy agendas, activities and cooperation mechanisms.

International collaborative research and researchers’ mobility worldwide are becoming key drivers of impact, and new major R&I players are emerging across the world. Over the last decade, the EU’s share of global expenditure in R&D has dropped from one fourth to one-fifth (European Commission, 2018a). Although certain fields of research have been broadly international since several decades, in recent years the worldwide landscape of research and innovation (R&I) has undergone substantial transformations, shifting towards an increasingly globalised and multipolar network of science and technology actors. In this context, the European Union's political priority to remain a major global actor through an R&I system that is open to the world requires cooperating more closely with international partner countries. The scope and interconnectivity of global societal challenges such as health or climate also call for global mobilisation of resources and coordination of activities. Furthermore, an increasing number of research fields require infrastructures which are

1 The cut-off date for the analysis data recorded in the Common Research Data Warehouse (CORDA), is 1/1/2019. Projects under Public-to-Public Partnerships, EIT’s Knowledge and Innovation Communities (KICs) and direct actions of the Joint Research Centre are not included except when explicitly mentioned. Box 2 still provides insights on participations in Public-to-Public partnerships. 2 Key data on the implementation of Horizon 2020 are publicly available on the Horizon 2020 Dashboard https://ec.europa.eu/info/funding- tenders/opportunities/portal/screen/opportunities/projects-results;programCode=H2020. 3 European Commission (2018a).

so sophisticated or costly that they exceed the capabilities of a single country, thereby leading to major multinational collaborations (European Commission, 2017). In this context, the EU needs to intensify its access to, and reap benefits from, the world’s best talents, expertise and resources in R&I.

Horizon 2020 is an essential vehicle for realising the objectives of the EU’s international cooperation strategy, as it enables collaboration between EU researchers and innovators and their best counterparts worldwide. Under Horizon 2020 targeted international cooperation actions shall be implemented on the basis of common priorities and mutual benefits, taking account of scientific and technological capabilities, market opportunities and expected impact. The objectives of international cooperation under Horizon 2020 apply in different ways depending on the international partner country or region, while areas for cooperation are identified on the basis of R&I capacities, market access opportunities, the contribution to international commitments, and the R&I framework conditions in place: . For EEA, EFTA and EU enlargement countries, the focus is on fostering integration into the European Research Area; . For European Neighbourhood Policy countries, the objective is to support a Common Knowledge and Innovation Space, including mobility for academics and capacity building; . For industrialised countries and emerging economies, the focus is on increasing competitiveness, jointly tackling global challenges and increasing participation in international value chains; . For developing countries, the emphasis is on promoting their sustainable development and addressing global societal challenges.

Evidence from the interim evaluation of Horizon 2020 (European Commission, 2017) but also from the FP7 ex- post evaluation of international cooperation (Farrell et al, 2015) showed the value of international cooperation not only to tackle global challenges and to support economic growth but also to deliver excellent research. As an illustration international collaboration increases the impact of scientific publications4: FP7 and Horizon 2020 peer-reviewed publications involving a contributor from at least one associated or third country are more cited than EU28 only publications and are cited at least three times more than the world average (European Commission, 2017).

Conclusions from the evaluation emphasised the need for intensifying international cooperation as a means to increase impact for the remainder of Horizon 2020 but also to seek alternative ways to increase participation of international partners in the longer term.

Overview of applications, participations and EU contribution to non-EU partners

EU Member States represent the vast majority of more than 650,000 applications and 100,000 participations in Horizon 2020 (89%) (Figure 1). A total of 4,700 distinct organisations from 16 associated countries and 108 non-associated third countries represent, respectively, 7% and 4% of participations so far (8% and 3% of applications). In terms of funding, associated countries represent 8% of the EU contribution provided through the programme (EUR 3 billion), while non-associated third countries represent only 0.6% of the funding (EUR 0.23 billion).

4 Craciun and Orosz (2018) recently reviewed the evidence supporting the benefits and costs of transnational collaborative partnerships in higher education. While the authors stress that not all assumed benefits of international cooperation in higher education are backed up by empirical evidence, in particular regarding the socio-cultural impact, they consider that there is a large evidence base showing that international collaborations in research activities result in more and better publications and patents.

Figure 1 Share of applications, participations and EU contribution by country group in Horizon 2020

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, Cut-off date 1 January 2019.

The distribution of participations by type of participants is similar between associated countries and EU countries (see Annex), with higher education institutions and private companies being the most represented for these countries (each of these category represents more than 30% of participations from associated countries and EU member states). Regarding non- associated third countries, there is a very large representation of higher education institutions (58% of their participations), which is mainly driven by the large participation of these countries in Marie Skłodowska-Curie Actions5 (MSCA). On the other hand, there is a low participation of private companies from non-associated countries (12%6 against 35% for EU member states).

Figure 2 presents the evolution of participations to collaborative projects7 between the 7th Framework Programme (FP7) and Horizon 2020. The figure shows that the share of participations from non-EU countries has shrunk between the two Programmes, especially for non-associated third countries. These results are in continuity with the results of the interim evaluation of Horizon 2020 (European Commission, 2017) and can be mainly attributed to the discontinuation of dedicated funding schemes for international cooperation, and changes in eligibility conditions for EU funding for certain third countries. As regards associated countries the slight decline in participations between both Programmes (from 8% of the participations and funding in FP7, to 7% of participations and 6% of the EU contribution in Horizon 2020) can be explained by the partial association of Switzerland during the first years of Horizon 2020.

5 See also Section 5. 6 16% without MSCA. 7 The focus here is on collaborative projects for comparability purpose with FP7. The figure excludes mono-beneficiary parts of the programme to which third countries are not automatically eligible to apply, and bottom-up parts. Collaborative projects in FP7 are all projects, with the exception of ERC and MSCA. In Horizon 2020, they exclude ERC, MSCA, SME instruments and Access to Risk Finance.

Figure 2 Evolution between FP7 and Horizon 2020 of participations and EU contribution in collaborative projects8

Note: Collaborative projects in FP7 exclude ERC and MSCA. Collaborative projects in Horizon 2020 exclude ERC, MSCA, SME instruments and Access to Risk Finance. Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, Cut-off date 1 January 2019.

Putting in perspective the national R&I intensity and the level of participation to the Programme of entities from individual associated countries and the most active (in terms of participations) non-associated third countries, some country groups emerge (Figure 3). Switzerland, Israel, and Norway perform fairly well in terms of R&D intensity and participate strongly in Horizon 2020. In this sense, they are key non-EU participants in the programme, both in terms of engagement in the programme and intensity of their national R&I efforts. Among S&T advanced non-associated third countries, the USA also shows the largest number of participations, mainly due to the strong involvement of US universities in Marie Skłodowska-Curie Actions (MSCA)9 targeting researchers’ mobility (see section 5) and the strong participation in the SC1 'Health, demographic change and wellbeing'. Other third countries with strong R&D intensities such as South Korea, Taiwan and Japan participate to a lesser extent in the programme.

8 The definition of country groups is based on the situation in Horizon 2020. 9 USA participations include 75% of participations in Marie Skłodowska-Curie Actions

Figure 3 R&D intensity and number of participations per year in Horizon 2020

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on Eurostat, OECD, World Bank and UNESCO for R&D intensity (last year available), and CORDA data for participations. *USA participations include 75% of participations in Marie Skłodowska-Curie Actions.

While participation is expected to correlate to some extent with the size of the countries10, this also shows a relationship in the degree of participation in Horizon 2020 with the level of investment made by countries in R&D as a percentage of GDP (i.e. R&D intensity), especially for associated countries.

More generally, this also illustrates that the highly competitive, excellence-driven nature of Horizon 2020 results in a differing level of engagement from third countries. Notably, those countries with less R&I endowments (illustrated here by lower R&D intensity) participate less11 than those with strong R&I support systems. On the other hand, as shown by the interim evaluation of Horizon 2020 dedicated support for policy, mobility and coordination activities have proven beneficial for some of the countries participating less in Horizon 2020, in particular associated countries with less-developed R&I capacities. As an illustration, for associated countries from the European Neighbourhood, the association to the programme has also contributed to the integration of their R&I systems within the European Research Area (European Commission, 2018).

10 When normalising by country population, Iceland and the Faroe Islands are the top associated countries in terms of participations per inhabitant, followed by Switzerland, Norway and Israel. Within non-associated third countries, New-Zealand, Australia and Canada have the highest number of participations per inhabitant. The US still presents a higher participation per inhabitant than most non-associated third countries. 11 The trend also holds when taking size into account, with Iceland, Norway, Switzerland and Israel showing the most participations per capita.

Countries associated to Horizon 2020

16 countries have signed an association agreement12 to Horizon 2020. Within these countries, Switzerland, Norway, Israel and Turkey are the countries submitting the highest number of applications (Figure 4), with respectively 32%, 21%, 18% and 12% of applications from this group of countries. In addition, Switzerland is the associated country with the highest share of high-quality proposals submitted (67%)13, followed by Iceland (61%) and Norway (60%). This is higher than the EU average (57%). In terms of success rate, the success rate for applications from associated countries (15.1%) is slightly higher than the success rate for EU applications (15%). Whereas the EFTA countries of Switzerland, Norway, Iceland all perform better than the EU average this is also the case for the Faroe Islands, Bosnia and Herzegovina and Tunisia. On the other hand, associated countries such as the candidate countries and most countries from the European Neighbourhood Policy tend to show lower performances than the EU average in terms of the quantity, quality and success of their applications.

Figure 4 Applications (left axis) and success rate (right axis) of associated third countries

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, Cut-off date 1 January 2019.

In line with the high share of high-quality proposals submitted, Switzerland is the most active associated country in terms of participations, with 2,808 participations, i.e. 37% of participations from associated countries (Figure 5). Norway, Israel and Turkey account respectively for 23%, 17% and 9% of participations from associated countries. In line with their application patterns, associated countries with the smallest participation (less than 1% of participations from associated countries) are Tunisia, Moldova, Georgia, Montenegro, Albania, Armenia and the Faroe Islands.

12 Of these, 11 countries have been fully associated since the start of the programme. Four agreements were signed in 2015 and 2016, while Switzerland was partially associated until the end of 2016 and is now associated to all parts of Horizon 2020. 13 Defined as those proposals which were evaluated as passing the quality threshold for a specific call.

Figure 5 Number of participations from associated countries in Horizon 2020 (% of all associated countries participations)

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, Cut-off date 1 January 2019.

Figure 6 presents the participations from associated countries in collaborative projects as a percentage of participations from all countries. Overall, the ranking of countries in terms of participations is similar between FP7 and Horizon 2020, with slight changes for some countries in terms of engagement in the Framework Programme. Within the most active countries, Switzerland, Israel and Turkey have experienced a decrease in their share of participations in collaborative projects between FP7 and Horizon 2020. The decrease for Switzerland can be mainly attributed to its partial association during the first years of Horizon 2020. On the other hand, Norway participates slightly more while Serbia has significantly increased its share in Horizon 2020 compared to FP7.

Figure 6 Participations from associated countries in collaborative projects in FP7 and Horizon 2020 (% of all participations in collaborative projects)

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, Cut-off date 1 January 2019.

Box 2 Public-to-public partnerships and multilateral initiatives

Non-EU countries are particularly well represented in public-to-public partnerships (P2P). Based on available data, associated countries account for 13% and non-associated countries for 6% of participations in P2Ps, significantly higher than the rest of the programme (see Figures 1 and 2). Moreover, with an EU budget of EUR 220 million, the PRIMA Art. 185 initiative partnering EU and south Mediterranean countries on R&I cooperation in agri-food and water issues, is expected to further boost international cooperation in P2P partnerships. With an annual contribution of around EUR 150-200 million, Horizon 2020 provides significant support to international cooperation through multilateral initiatives, notably in the areas of health, environment, food and energy.

There is a strong similarity in the distribution of participations across the different programme parts between EU countries and associated countries (Figure 7). A fifth of participations are within the MSCA schemes targetting individual researchers. The main differences are observed in European Research Council projects, where the share of participations from associated countries is significantly larger than for EU participants (9.5% against 5.5%, corresponding to 36% of EU investment for associated countries) and in Societal Challenge 4 (“Smart, Green and Integrated Transport”) where it is lower (5.5% against 8.8% for EU participants).

Figure 7 Number of participations from associated countries by programme part (% of participations in the country group)

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, Cut-off date 1 January 2019. Definition of acronyms in Annex.

Looking closer at collaboration networks within Horizon 2020, Figure 8 maps collaborations (based on joint participations in projects) within organisations located in EU and associated countries. This shows two types of connections: (i) the single strongest connection of each country to another country, and (ii) the top 40 strongest connections within the network. The size of the nodes is proportional to the centrality of the countries.

From this analysis it comes out that Switzerland occupies a very central14 position in the network of collaborations amongst participants in Horizon 2020, next to other EU28 countries such as Sweden, Greece and Austria. Figure 8 also shows which specific countries are the most frequent entry gates to Horizon 2020 for associated countries. For example, Israel, Georgia and Iceland collaborate frequently with Germany, while Turkey shows a strong collaborative relationship with Spain. Norway displays close connections with Spain, the UK, France and Germany. Italy is a frequent collaborator for Serbia, Albania, North Macedonia, Moldova and Bosnia and Herzegovina, while France has strong connections with Ukraine and Tunisia.

Figure 8 Horizon 2020 network of projects’ participations – Focus on EU28 and Associated Countries

Note: This graph represents the backbone of Horizon 2020. Nodes are countries of the participating organisations, and links represent strong15 connections based on Horizon 2020 projects. Non-associated third countries are not represented on the graph. Source: European Commission (2018c).

Participants from associated countries are on average less central in the network of Horizon 2020 collaborations than EU Member States, but they are characterised by a high degree of heterogeneity, with the least central country (Armenia) being ranked 83rd in terms of centrality amongst all countries participating in Horizon 2020 and the most central country (Switzerland) being in the 9th place16 (Figure 9). After Switzerland, the most central countries are Norway (14th), Israel (20th) and Turkey (24th). These four countries have been particularly stable in terms of centrality in the network since the 6th Framework Programme (FP6). They are followed by Serbia (33th), Ukraine (35th) and Iceland (36th), which have also been stable in this ranking since FP6. Participants from North Macedonia (47th), Moldova (54th), Faroe Islands (65th), Albania (66th) and Bosnia and Herzegovina (67th) have improved their position in the collaboration network since FP6.

14 Centrality can be defined as the importance of a country in the network of collaboration between Horizon 2020 participants. This importance as such can have different meanings, hence different definitions, with the most straightforward definition being based on the number of connections of a country’s participants in the whole network. 15 The links displayed on this graph with N actors combines the N-1 links of a maximum spanning tree (MST) and the N-1 strongest links of the overall network. The MST represents the backbone of a weighted network and is based on three rules. First, it keeps only N-1 links from a network with N actors. Second, rule #1 should be satisfied while keeping the strongest links. Third, rule #1 and #2 should be satisfied without creating any isolate in the network. 16 Balland and Ravet (2018).

Figure 9 Ranking of associated countries in terms of centrality (within all countries)

Source: Balland and Ravet (2018).

Non-associated third countries

With applicants from 163 non-associated third countries so far, Horizon 2020 demonstrates a very broad international outreach. The USA gathers about 30% of these applications (Figure 10), followed by China (8%), Canada (6%) and Australia (5%). Overall the top-20 applicant countries gather 81% of these applications whereas 45 countries submitted only 1 or 2 applications. The large share of applications from the USA is mainly explained by a high number of MSCA applications for individual researchers that involve US organisations. The level of quality of the applications from these countries is also particualrly high: almost 80% of applications submitted by US participants were evaluated as high quality. This share is around 70% for many of the most active non-associated third countries, compared to less than 60% for associated countries and EU Member States. In the same line the success rate for applications from non- associated third countries is higher than for other country groups (18% compared to 15% for EU applications). Still, some countries submit relatively less high-quality proposals than EU28 (57% of applications in high-quality proposals).

Figure 10 Number of applications (left axis) and success rate (right axis) of non- associated third countries

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, Cut-off date 1 January 2019.

Regarding participations, a total of 108 non-associated third countries are represented in Horizon 2020 projects. With over 1,100 participations so far, the USA accounts for 28% of participations from non- associated third countries (Figure 11, with more data for all non-associated countries in Annex). The USA is followed by China (9% of participations from non-associated third countries), Canada (6%), Australia (5%), South Africa (4%) and Brazil (4%). Overall the top-20 participant non-associated third countries gather 81% of these participations, with a low level of participation for many developing economies.

Figure 11 Participations from non-associated third countries in Horizon 2020

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, Cut-off date 1 January 2019.

Compared to FP7, there is a clear drop in the participation of non-associated third countries in collaborative projects17. Figure 12 shows that participations from most top non-associated third countries in collaborative projects represent a lower share of participations in Horizon 2020 compared to FP7, with Russia, Brazil, India, Japan, Mexico Morocco and Egypt exhibiting the largest relative decrease. As stated in Section 3, this can be mainly explained by the discontinuation of dedicated international cooperation schemes that existed in FP7 and the change in eligibility conditions for funding participants from Brazil, Russia, India, China and Mexico between FP7 and Horizon 2020. To counterbalance this trend, international cooperation flagships were introduced in the Work Programme 2018-2020 of Horizon 2020 as a measure to strengthen strategic cooperation with key third partners on areas of mutual benefit.

The preliminary results of the 2018 calls exhibit an increase in the participations of non-associated third countries to collaborative projects to 3.6% in 2018 (based on around 40% of signed projects) as compared to 2.4% average for years 2014-2017. There is also a significant increase in non-associated countries’ contributions to entities in collaborative actions (from €26 million in 2014 to €113 million in 201718), due to efforts to extend dedicated Horizon 2020 co-funding mechanisms in international partner countries.

Figure 12 Participations from top-20 non-associated third countries in collaborative projects in FP7 and Horizon 2020 (% of all participations in collaborative projects)

Source : DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, Cut-off date 1 January 2019.

Figure 13 presents participations from non-associated third countries by Horizon 2020 programme part. About half of these participations take place in MSCA projects (52.1% of participations while MSCA represent 18.2% of EU participations). This corresponds to 11% of all participations and 0.1% of the EU investment in MSCA. This is driven by the USA, which constitutes 40% of these participations. Projects in the Societal Challenge 1 “Health, demographic change and wellbeing” are also particularly well represented (9% of participations from non-associated third countries). This can be partly explained by the fact that US participants are eligible to receive EU funding under this Societal Challenge. Other programme parts with a large representation of participations from non-associated third countries are the Societal Challenge 2 (“Food security, sustainable agriculture and forestry, marine and maritime and inland water research, and the Bioeconomy”, with 9.3% of participations) and Societal Challenge 5 (“Climate action, environment, resource efficiency and raw materials”, with 7.1% of participations). Compared to EU participations, non-associated third countries participate significantly less in the programme parts dedicated to Leadership in Enabling and Industrial Technologies (LEIT), to Societal Challenge 4 (“Smart, green and integrated transport”) and Societal Challenge 3 (“Secure, clean and efficient energy”).

17 As in Section 3, the comparison with FP7 is made on the basis of collaborative projects for comparability purpose. 18 Overall, the own contribution from non-associated countries has been about EUR 300 million so far.

Figure 13 Participations from non-associated third countries by programme part (% of participations in the country group)

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, based on CORDA data, Cut-off date 1 January 2019. Definition of acronyms in Annex.

Participants from non-associated third countries are, on average, less central than participants from other country groups in the network of collaborations between participants (European Commission, 2018c). Similar to associated countries, their most frequent partners are participants from large EU countries such as the UK, Germany, France, Italy and Spain. As shown in Figure 14, US participants are still highly ranked in terms of centrality (23rd most central country in Horizon 2020) and this has been reinforced over time since FP6. The picture is mixed among BRIC countries, with China maintaining its second most central position within non- associated third countries, while the position of Russia and India has dropped in this ranking over time.

Figure 14 Ranking of top non-associated third countries in terms of centrality (within all countries)

Source: Balland and Ravet (2018). Note: top 15 most central countries In Horizon 2020 plus India.

Overall messages

With participants from 124 non-EU countries, Horizon 2020 demonstrates a very broad international outreach attracting talent from around the world in particular from higher education organisations.

The monitoring flash overall shows a certain level of heterogeneity in the cooperation patterns with third countries, reflecting the diversity of objectives and benefits pursued with international cooperation: . Horizon 2020 is benefitting from excellence worldwide for increasing competitiveness, jointly tackling global challenges and increasing participation in international value chains through the involvement of participants from 108 non-associated third countries. Following a decrease of participation from non-associated third countries in collaborative projects since FP7, the international cooperation flagships of the Horizon 2020 Work Programme 2018-2020 are starting to partly reverse this picture, with encouraging preliminary results: based on slighly less than half of projects from 2018 calls, the participation share of non-associated third countries is around 50% higher than the average during 2014-2017. Moreover, cooperation with both associated and non-associated countries has been reinforced in Public-Public Partnerships and multilateral initiatives. Most of the collaborations are with a set of countries with advanced R&I capabilities in particular through researchers mobility schemes

such as MSCA but also through specific projects and multilateral initiatives to support sustainable development and addressing global societal challenges including with developing economies. Countries with strong R&I performance such as Switzerland, Norway and Israel are the most active associated countries, while almost one third of participations from non-associated third countries are from the USA (partly due to a large US participation in MSCA schemes).

. Horizon 2020 also contributes to the integration of R&I systems in the ERA for countries which have a relative lack of R&I capacity as well as supports a Common Knowledge and Innovation Space, including through mobility for researchers. The participation of countries with less-advanced R&I capacities can often prove challenging as shown by the under-EU average performance in terms of the quantity, quality and success of applications of these countries.

While the international dimension of Horizon 2020 has been reinforced for the remaining years of Horizon 2020 through targeted actions in the last Work Programme, the identified trends also call for an intensification of international cooperation activities in Horizon Europe. The proposal made for Horizon Europe based on its impact assessment (European Commission, 2018a) is expected to enhance the excellence and impact of the programme, allowing EU participants to collaborate with the best minds in the world for increased excellence and competitiveness, for effectively tackling global challenges and for implementing global commitments. In addition to fostering the creation and diffusion of high-quality knowledge in the EU, it would also give to the EU a higher influence in shaping global R&I systems. It would enhance the EU’s leading role in setting the policy agenda, in particular for addressing common challenges and for achieving the Sustainable Development Goals. Horizon Europe is expected to be an effective instrument in Europe's efforts to harness globalisation by removing barriers to innovation and by establishing fairer framework conditions with international partners.

Towards Horizon Europe – based on the Commission proposal

Horizon Europe should promote and integrate cooperation with third countries and international organisations and initiatives based on common interest, mutual benefit and global commitments to implement the UN Sustainable Development Goals. International cooperation should aim to strengthen the Union's research and innovation excellence, attractiveness and economic and industrial competitiveness, to tackle global challenges, as embodied in the UN SDGs, and to support the Union's external policies. An approach of general opening for international participation and targeted international cooperation actions should be followed, including through appropriate eligibility for funding of entities established in low to middle income countries. At the same time, association of third countries to the Programme should be promoted.

Overall, as summarised in the impact assessment of Horizon Europe, the programme is expected to: . Extend its openness to association of third countries to make cooperation and co-funding as smooth as possible;. . Continue its general opening for international participation, for entities from both industrialised and from developing countries, and continue its funding of entities from low-mid income countries and only exceptionally of entities from industrialised countries; . Launch targeted actions to pursue strategic international cooperation in line with EU priorities; and . Be more proactive in seeking synergies with other Union programmes, including the External Instrument.

References Balland, P.A., and Ravet, J. (2018). Dynamic Network Analysis of the EU R&I Framework Programme. European Commission report. Craciun, D. and Orosz, K. (2018). Benefits and costs of transnational collaborative partnerships in higher education. EENE Analytical Report No.36 prepared for the European Commission. Farrell M., Kalpazidou Schmidt E., Mourzelas M., Warrington B., Wood J. (2015), Ex-post Evaluation of International Cooperation Activities of the Seventh Framework Programme’s Capacities Programme, Report for the European Commission European Commission (2017 ). Interim Evaluation of Horizon 2020. Staff Working Document. SWD(2017)220. European Commission (2018a). Impact Assessment of Horizon Europe. Staff Working Document. SWD(2018)307. European Commission (2018b). Country Participation. From Horizon 2020 to Horizon Europe - Monitoring Flash #1. European Commission (2018c). Dynamic Network Analysis. From Horizon 2020 to Horizon Europe - Monitoring Flash #2.

Annex Table 1 Programmes parts in Horizon 2020

ACRONYM PROGRAMME PART ERC European Research Council MSCA Marie Skłodowska-Curie actions FET Future and Emerging Technologies RI Research infrastructures (including e-infrastructure) LEIT Leadership in enabling & industrial technologies ARF Access to risk finance Innovation in SMEs Innovation in SMEs FTI Fast Track to Innovation SC1 Health, demographic change & wellbeing SC2 Food security, sustainable agriculture and forestry, SC3 Secure, clean & efficient energy SC4 Smart, green & integrated transport SC5 Climate action, environment, resource efficiency & raw materials SC6 Inclusive, innovative & reflective societies SC7 Secure societies SEWP Spreading excellence & widening participation SWAFS Science with and for society Euratom Euratom

Table 2 Participations by type of participant

ASSOCIATED NON-ASSOCIATED WHOLE PROGRAMME EU28 TOTAL COUNTRIES THIRD COUNTRIES Higher education 2918 38% 2392 58% 29241 32% 34551 34% Private sector 2520 33% 477 12% 32190 35% 35187 33% Public bodies 580 8% 266 7% 5055 6% 5901 6% Research organisations 1298 17% 666 16% 20044 22% 22008 21% Other 278 4% 288 7% 5262 6% 5828 6% Total 7594 100% 4089 100% 91792 100% 103475 100% ASSOCIATED NON-ASSOCIATED WITHOUT MSCA EU28 TOTAL COUNTRIES THIRD COUNTRIES Higher education 2099 34% 731 38% 20036 27% 22866 28% Private sector 2202 36% 313 16% 28474 38% 30989 37% Public bodies 531 9% 223 11% 4777 6% 5531 7% Research organisations 1121 18% 462 24% 16927 23% 18510 22% Other 248 4% 220 11% 4844 6% 5312 6% Total 6201 100% 1949 100% 75058 100% 83208 100%

Table 3 Participations and EU investment through Horizon 2020 – associated countries NR OF HORIZON 2020 PARTICIPATIONS IN SIGNED HORIZON GRANTS BY % OF TOTAL 2020 COUNTRY OF HORIZON 2020 INVESTMENT % OF TOTAL HORIZON ASSOCIATED COUNTRIES BENEFICIARY PARTICIPATIONS (EUR MILLION) 2020 INVESTMENT Albania 27 0.03% 2.4 0.01% Armenia 26 0.03% 1.2 0.00% Bosnia and Herzegovina 63 0.06% 5.5 0.01% Faroe Islands 18 0.02% 2.8 0.01% Georgia 30 0.03% 2.7 0.01% Iceland 243 0.23% 85 0.22% Israel 1255 1.21% 732.6 1.88% Moldova (Republic of) 56 0.05% 5.1 0.01% Montenegro 29 0.03% 1.6 0.00% North Macedonia 66 0.06% 7.3 0.02% Norway 1746 1.69% 800.3 2.06% Serbia 339 0.33% 72.3 0.19% Switzerland 2808 2.71% 1128.9 2.90% Tunisia 61 0.06% 7.7 0.02% Turkey 652 0.63% 144.7 0.37% Ukraine 175 0.17% 20.8 0.05% Total 7594 7.3% 3020.9 7.8% Source: European Commission, DG RTD, based on CORDA data, Cut-off date 1 January 2019.

Table 4 Participations and EU investment through Horizon 2020 – Non-associated third countries NR OF HORIZON 2020 PARTICIPATIONS IN HORIZON SIGNED GRANTS BY % OF TOTAL 2020 % OF TOTAL NON-ASSOCIATED THIRD COUNTRY OF HORIZON 2020 INVESTMENT HORIZON 2020 COUNTRIES BENEFICIARY PARTICIPATIONS (EUR MILLION) INVESTMENT Afghanistan 2 0.00% 0.8 0.00% Algeria 12 0.01% 0.7 0.00% Angola 1 0.00% 0 0.00% Anguilla 4 0.00% 0.9 0.00% Argentina 130 0.13% 5.3 0.01% Australia 215 0.21% 4.9 0.01% Azerbaijan 10 0.01% 0.5 0.00% Bangladesh 3 0.00% 0.8 0.00% Belarus 43 0.04% 2.3 0.01% Benin 3 0.00% 0.8 0.00% Bolivia 5 0.00% 0.2 0.00% Botswana 3 0.00% 0.3 0.00% Brazil 162 0.16% 9.7 0.02% British Virgin Islands 1 0.00% 0.2 0.00% Burkina Faso 13 0.01% 5.6 0.01% Burundi 2 0.00% 0.1 0.00% Cambodia 3 0.00% 0.5 0.00% Cameroon 9 0.01% 0.7 0.00% Canada 240 0.23% 4.7 0.01% Cape Verde 9 0.01% 0.3 0.00% Chile 91 0.09% 4.2 0.01% China (People's Republic of) 362 0.35% 3.1 0.01%

NR OF HORIZON 2020 PARTICIPATIONS IN HORIZON SIGNED GRANTS BY % OF TOTAL 2020 % OF TOTAL NON-ASSOCIATED THIRD COUNTRY OF HORIZON 2020 INVESTMENT HORIZON 2020 COUNTRIES BENEFICIARY PARTICIPATIONS (EUR MILLION) INVESTMENT Colombia 53 0.05% 3.8 0.01% Costa Rica 16 0.02% 0.9 0.00% Cote d'Ivoire 4 0.00% 0.2 0.00% Cuba 15 0.01% 0.6 0.00% Dominican Republic 1 0.00% 0 0.00% Ecuador 17 0.02% 1.3 0.00% Egypt 41 0.04% 3.5 0.01% Ethiopia 27 0.03% 2.5 0.01% French Polynesia 4 0.00% 0 0.00% Gabon 2 0.00% 0.8 0.00% Ghana 24 0.02% 4.4 0.01% Gibraltar 5 0.00% 2.9 0.01% Greenland 9 0.01% 0.7 0.00% Grenada 1 0.00% 0 0.00% Guatemala 3 0.00% 0.2 0.00% Hong Kong 19 0.02% 0.6 0.00% India 96 0.09% 2.4 0.01% Indonesia 20 0.02% 1.2 0.00% Iran (Islamic Republic of) 13 0.01% 0.6 0.00% Iraq 4 0.00% 0.3 0.00% Jamaica 6 0.01% 1.4 0.00% Japan 124 0.12% 2.8 0.01% Jersey 1 0.00% 0 0.00% Jordan 15 0.01% 5.4 0.01% Kazakhstan 8 0.01% 0.3 0.00% Kenya 65 0.06% 9.8 0.03% Kosovo * UN resolution 11 0.01% 1.1 0.00% Kyrgyzstan 8 0.01% 0.5 0.00% Lao (People's Democratic Republic) 2 0.00% 0 0.00% Lebanon 21 0.02% 1.6 0.00% Lesotho 1 0.00% 0 0.00% Liberia 1 0.00% 0.1 0.00% Libya 1 0.00% 0 0.00% Liechtenstein 5 0.00% 0 0.00% Madagascar 7 0.01% 0.1 0.00% Malawi 9 0.01% 2.2 0.01% Malaysia 29 0.03% 1.1 0.00% Mali 4 0.00% 0.2 0.00% Mauritania 1 0.00% 0.1 0.00% Mauritius 2 0.00% 0 0.00% Mexico 63 0.06% 0.9 0.00% Mongolia 3 0.00% 0 0.00% Morocco 59 0.06% 4.1 0.01% Mozambique 8 0.01% 1 0.00% Myanmar 1 0.00% 0.1 0.00% Namibia 8 0.01% 0.6 0.00% Nepal 5 0.00% 1.2 0.00% New Caledonia 8 0.01% 0.3 0.00%

NR OF HORIZON 2020 PARTICIPATIONS IN HORIZON SIGNED GRANTS BY % OF TOTAL 2020 % OF TOTAL NON-ASSOCIATED THIRD COUNTRY OF HORIZON 2020 INVESTMENT HORIZON 2020 COUNTRIES BENEFICIARY PARTICIPATIONS (EUR MILLION) INVESTMENT New Zealand 46 0.04% 1.8 0.00% Nicaragua 1 0.00% 0.3 0.00% Niger 3 0.00% 0.8 0.00% Nigeria 15 0.01% 1.3 0.00% Pakistan 13 0.01% 1.2 0.00% Palestine 8 0.01% 0.3 0.00% Panama 1 0.00% 0.1 0.00% Paraguay 5 0.00% 0.3 0.00% Peru 22 0.02% 1.8 0.00% Philippines 5 0.00% 0.1 0.00% Qatar 3 0.00% 0 0.00% Russian Federation 124 0.12% 3 0.01% Rwanda 7 0.01% 1.3 0.00% Sao Tome and Principe 1 0.00% 0 0.00% Saudi Arabia 5 0.00% 0 0.00% Senegal 36 0.03% 3.1 0.01% Seychelles 1 0.00% 0.1 0.00% Sierra Leone 3 0.00% 8.1 0.02% Singapore 19 0.02% 0.2 0.00% South Africa 174 0.17% 27.5 0.07% South Korea 67 0.06% 0.7 0.00% Sri Lanka 6 0.01% 1.1 0.00% Swaziland 4 0.00% 0.3 0.00% Taiwan 46 0.04% 0.7 0.00% Tajikistan 3 0.00% 0.4 0.00% Tanzania (United Republic of) 22 0.02% 6 0.02% Thailand 29 0.03% 1.1 0.00% Togo 2 0.00% 0.1 0.00% Turkmenistan 1 0.00% 0.1 0.00% Uganda 29 0.03% 6.2 0.02% United Arab Emirates 2 0.00% 0 0.00% United States 1147 1.11% 45.1 0.12% Uruguay 20 0.02% 2.8 0.01% Uzbekistan 4 0.00% 0.1 0.00% Venezuela 6 0.01% 0 0.00% Viet Nam 30 0.03% 1.7 0.00% Yemen 1 0.00% 0.1 0.00% Zambia 5 0.00% 1.4 0.00% Total 4089 4.0% 227.6 0.6% Source: European Commission, DG RTD, based on CORDA data, Cut-off date 1 January 2019.

Table 5 Participations and EU investment by programme part

PARTICIPATIONS HORIZON 2020 INVESTMENT % ALL % TOTAL % TOTAL PART. NON- % ALL INVESTMENTS NON- INVESTMENTS IN ASSOCIATED PART. IN IN ASSOCIATED IN

PROGRAMM ASSOCIATED PROG. THIRD PROG. ASSOCIATED PROGRAMME THIRD PROGRAMME © European Union, 2020 E PARTS COUNTRIES PART COUNTRIES PART COUNTRIES PART COUNTRIES PART ERC 724 12.4% 45 0.8% 1085.7 13.9% 9.7 0.1% MSCA 1393 6.9% 2140 10.6% 275.9 7.1% 2.1 0.1% FET 248 8.5% 20 0.7% 165.3 11.5% 1.4 0.1% RI 428 8.7% 143 2.9% 122 7.8% 19.2 1.2% LEIT-ICT 773 6.5% 196 1.6% 244.7 5.5% 12.5 0.3% LEIT-NMBP 452 6.8% 101 1.5% 156 6.0% 7.4 0.3% LEIT-SPACE 117 5.5% 52 2.4% 22.6 3.7% 4.2 0.7% Innovation in SMEs 339 12.3% 8 0.3% 65.5 13.3% 0.2 0.0% FTI 55 8.5% 1 0.2% 19 7.1% 0.3 0.1% SC1 Health 509 6.5% 380 4.9% 129.4 3.8% 90.4 2.6% SC2 Food 538 7.8% 327 4.7% 164.1 8.4% 23.9 1.2% SC3 Energy 617 7.7% 52 0.6% 219 7.3% 5.4 0.2% SC4 Transport 416 4.8% 82 1.0% 107.8 3.7% 2 0.1% SC5 Climate 378 7.1% 292 5.5% 97.9 6.2% 26.3 1.7% SC6 Societies 195 7.3% 174 6.5% 38.5 6.0% 17.6 2.7% SC7 Security 210 7.9% 18 0.7% 60.8 7.5% 1 0.1% SEWP 39 4.6% 2 0.2% 20.6 3.9% 0 0.0% SWAFS 115 8.3% 37 2.7% 15.4 5.7% 3.1 1.1% Euratom 44 3.9% 19 1.7% 10 1.6% 0.4 0.1% Total 7594 7.3% 4089 4.0% 3020.9 7.8% 227.6 0.6% Source: European Commission, DG RTD, based on CORDA data, Cut-off date 1 January 2019.

Other monitoring flash reports available here: https://ec.europa.eu/info/publications/horizon-2020-monitoring-flash_en

#1 Country Participation #2 Dynamic Network Analysis #4 Patents in FP #5 Sustainable Development Goals

4 PATENTS IN RAMEOR PROGRAMME

by Martina Kadunc, Daniel Neicu, Šimon Trlifaj, Abakouy Hecham

FROM HORIZON 2020 TO HORIZON EUROPE MONITORING FLASH PATENTS IN FP

August 2020 This Monitoring Flash analyses the self-reported patenting activity of beneficiaries of the European Framework Programmes for Research and Innovation (FP) between 2009 and 2018; FP7 (2007-2014) and Horizon 2020 (2014-2020). It sheds light on the type of organisations that own FP patents, the types of inventions reported and their estimated market value. FROM FP7 TO HORIZON 2020 Key overview data

10 920 2 776 75% 622 (6%) self-reported patents linked self-reported inventions of patents owned by EU of self-reported patents to more than 50,000 FP protected by 10 920 patents organisations based on 50% related to COVID-19 projects ownership rule pandemic SELF-REPORTED INVENTIONS FROM FP PROJECTS Key overall messages . 97% of the FP inventions between 2009 and 2018 are results of FP7 projects. This indicates a considerable time-lag between FP activities and potential exploitation through patents. In the future, time-lag could be reduced through targeted interventions. . The FP self-reported patents have in general higher average IPBI estimated market value which indicates the Added Value of the FP. . The majority of FP self-reported inventions are related to the health sector in areas such as biotechnology, pharmaceuticals or organic chemistry, a limited number is related to environmental technologies. This could reflect the policy priorities of the past (in FP7) and a different picture is likely to emerge for Horizon 2020 and Horizon Europe, which put a more explicit focus on climate action . FP self-reported patents are intended to be largely exploited at home in Europe and the United States. The majority of the innovations are owned by European organisations. More than half of them are owned by SMEs, but in comparison with the EU and the World this share is low. A comparatively large share of patents is owned by large and very large organisations. . Patenting is only one of possible outputs of FP projects but remains the most widely used indicator of innovation. FP patent data face significant quality challenges which limits the analytical and policy conclusions of this Monitoring Flash.

Introduction

The EU Research and Innovation Framework Programmes (FP) aim to foster all forms of innovation, and particularly those that can tackle our biggest environmental, economic and societal challenges. By supporting pan-European collaborative research and innovation, small and medium-sized firms and individual researchers, the FPs strive to speed up the development of the technologies and innovations that will underpin tomorrow’s businesses and help European companies to grow1. Once an innovation is developed, the inventor seeks to protect it either with a patent or other Intellectual Property Right (IPR), or by secrecy (Cohen et al., 2000; Hall et al., 2013). Whereas patents present only a small part of what results from R&I activities they are currently still one of the most widely used indicators of innovation.

Patents incentivise inventors to pursue R&I activity, codify newly created knowledge and help its diffusion2. Existing studies show, that public funding can increase patenting activity of firms (Howell, 2017; Wildmann, 2017), especially of small firms (Bronzini & Piselli, 2016). The Horizon 2020 Interim Evaluation gives some insights into the patenting activity of FP beneficiaries3. It suggests that the patents produced in the FPs are of higher quality and likely commercial value than patents produced elsewhere.

This Monitoring Flash sheds further light on patents and patenting activity of innovators benefiting from the two latest Framework Programmes – FP7 and Horizon 2020, based on self-reported project results. The analysis covers self-reported patents from more than 50,000 FP7 and Horizon 2020 projects funded until 2019. The data is stored in the Common Research Data Warehouse (CORDA), an internal database maintained by DG Research and Innovation. The analysis is based on a pilot matching of CORDA data with ORBIS Intellectual Property database4. This pilot activity tested the possibilities of gaining new analytical insights by enhancing internal EC data on patents with an external dataset.

The Monitoring Flash was prepared by the European Commission services, however it reflects the views only of the authors, and the Commission cannot be held responsible for any use which may be made of the information contained therein.

1 Knowledge transfer is one of the means through which the Framework Programmes aim to increase the innovative capability of firms, reinforce competitiveness and spur economic growth (European Commission, 2017). 2 Other effects of patents are ambiguous and remain to be discussed in academic literature, see Pisano & Teece, 2007; Pries & Guild, 2011; Penin & Neicu, 2018. 3 European Commission (2017) Interim Evaluation of Horizon 2020, Staff Working Document SWD(2017)2020 and PPMI, ‘Assessment of the Union Added Value and the Economic Impact of the EU Framework Programmes (FP7 and Horizon 2020) 4 https://www.bvdinfo.com/en-gb/our-products/data/international/orbis-intellectual-property

Framework Programme patents How many inventions result from Framework Programme projects? There are currently 2 776 inventions (patent families)5 self-reported by FP7 and Horizon 2020 beneficiaries protected by 10 920 patents worldwide. The vast majority (97%) are results of FP7 projects.

Due to the considerable time lag between achieving research and innovation results, applying for patents and the length of the patenting procedure at different patenting offices (European Commission, 2017: p. 132) the expected coverage of FP self-reported patents remains low. Based on the data, the average project starts reporting patents 16 months after the project start. The maximum observed time for reporting a patent was 8 years after the project start. Whereas much of the FP7 patents that can be obtained through self-reporting have probably been collected by now, the number of Horizon 2020 reported patents is expected to increase significantly in the coming years. For instance, as seen in Figure 1 below, the first Horizon 2020 patents started to be reported only in 2016, two years after the official start of Horizon 2020 programme. The first Horizon 2020 patents come from the LEIT-ICT, ERC proof of concept and the Energy part of the programme.

Figure 1 Number of self-reported FP inventions (left axis) and number of all reported inventions in the World (right axis) by publication year

600 1.600.000

1.400.000 500 1.200.000 400 1.000.000

300 800.000

600.000 200 400.000 100 200.000

0 0 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018

FP7 patent families H2020 patent families WORLD patent families w/o CN

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit based on ORBIS Intellectual Property (IP), CORDA.

The quality of self-reported patents remains low. After applying several steps of quality assurance process, only 42% of the self-reported patents remained for further analysis. The majority of self-reported patents were excluded due to the background rather than foreground nature of the patents. Foreground patents are results of the FP activity6, whereas background patents are an input, a starting point for the FP activity. Around 50% of the self-reported patents were identified as background patents and some were very old, even from the 19th century. In addition, due to the voluntary reporting during the project lifetime, a large share of patents linked to FP projects remains unreported. The discovery of these is the object of a series of ongoing projects of the European Commission.

5 A patent family is a set of patents that protect the same invention disclosed by the same inventor in different countries. The number of patent families hence indicates the number of distinct inventions protected. 6 The patent application date or first priority date is later than one year after the start of the project. In doing so, we assume that it takes FP beneficiaries at least one year from starting the project to applying for a patent with one of the worldwide patent offices. See the methodological Annex for more information.

The present low patent data quality and tracing challenges indicate that the European Commission should increase its efforts in ensuring appropriate and robust monitoring of patenting activity of FP beneficiaries. This improve evidence base for future policy making. The current data limitations should be taken into account when reading the FP patent analysis that follows.

What type of Framework Programme inventions are patented? The majority of the FP self-reported inventions (patent families) are patented in health-related areas such as biotechnology, pharmaceuticals, organic chemistry or medical technology. Only a limited number of inventions relate to environmental technology.

The highest share of FP self-reported inventions (patent families) is related to biotechnology7 (14% of all self- reported inventions). This is 9 times higher than the worldwide average (1.5% of world inventions are in biotechnology). Pharmaceutical inventions follow with around 9% of FP inventions, more than 4 times the worldwide average (2% of world inventions are in this class) and 3 times more than inventions registered in the EU in this class (2.7% of all EPO and EU28 inventions). The FP has a relatively high share of inventions also in the Analysis of the biological materials and Micro-structural and nano-technologies; the share of FP inventions is these classes is more than ten times higher than the world average.

On the other hand, based on the data at hand, the Framework Programmes produce proportionately less inventions in the IT methods for management, as well as in Audio-visual technology classes, among others. If some of these results are in line with how Europe is performing – measured by the percentage of EPO and EU28 inventions in each of these classes – others seem to point to a lesser focus of the FP compared to the overall picture in Europe. This is especially the case for the transport and telecommunications areas. Environmental technology inventions are also under-represented in FP7 and Horizon 2020, but this follows a general European trend seen by lower percentages of inventions in this field registered at the EPO and EU28 patent offices if compared to the World.

Bearing in mind the highlighted data limitations, such an analysis hints to the fact that the FP is highly specialised in health and medicine-related technological areas also when compared to the World and to the European average. In some fields, this follows the European comparative areas of specialisation including biotechnology, organic and macromolecular chemistry and nano-technology. However, in other fields, the FP seems to have some niche specialisations (not observed for Europe) including analysis of bio-materials. The analysis also indicates that the past FPs contributed little to the environmental technology inventions. This is a particularly important message, given the future policy ambition of the new Commission for Europe to become the first climate-neutral continent by 2050. There are several potential explanations for such observation: the WIPO technology class ‘environmental technologies’ is a too narrow indicator for climate related patents8; the past investment in climate related research and innovation was limited (in FP7) or not focused on producing innovations; or the time-lags for development of climate related technologies are much longer than those in health-related areas and hence not captured by self-reporting during the project lifetime. With the increased role of EU R&I in enabling the transition to climate-neutrality, it is important that FP project results and impacts are indentified in a more comprehensive and systematic way compared to the current status.

7 Note that WIPO technology classes are counted only for the main patent of each FP foreground patent family, due to data constraints. Worldwide figures are, nevertheless, at patent-level, rather than patent family (invention) level. Given that the patents covering an invention are very similar, one can assume that they are registered in the same WIPO class. 8 The WIPO environmental technology classification does not cover all relevant ‘green patents’. The OECD developed Green Patents methodology to find patents in environment-related technologies which draws on more than 200,000 classification symbols. The search strategies encompass a broad spectrum of technologies related to environmental pollution, water scarcity, climate change mitigation. A comparable methodology would need to be applied to appropriately asses the FP contribution to green innovation.

Figure 2 Technological specialisation index of FP inventions, EPO published in 2009-2018 (Worldwide=1)

14 FP families EU families (EPO + EU28) 12 WORLD 10

0 Electrical machinery, apparatus, energy apparatus, machinery, Electrical technology Audio-visual Telecommunications Digital communication communication processes Basic Computer technology for management methods IT Semiconductors Optics Measurement materials biological of Analysis Control technology Medical Organic fine chemistry Biotechnology Pharmaceuticals polymers chemistry, Macromolecular chemistry Food chemistry materials Basic metallurgy Materials, coating technology, Surface Micro-structural and nano-technology engineering Chemical technology Environmental Handling Machine tools turbines pumps, Engines, machines paper and Textile machines special Other apparatus and processes Thermal Mechanical elements Transport games Furniture, Other consumer goods Civil engineering

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit based on ORBIS Intellectual Property (IP), CORDA. Notes: Based on main patents a) Values are normalised so that worldwide percentage of patents in each WIPO technology class equals 1. A value of 2 indicates a percentage (of FP or EU patent families) twice as high as the worldwide percentage of patents in that class.

Have the Framework Programmes supported inventions relevant to the COVID-19? 622 (6%) of self-reported FP patents between 2009 – 2018 seem relevant to the current fight against the COVID-19 pandemic9. 68 (11%) of these patents relate to coronavirus or respiratory drugs and 121 (19%) to personal protective equipment including masks and gowns.

Figure 3 FP COVID-19 related self-reported patens published in 2009-2018, by type

11% 19%

662 FP patents

70%

CORONAVIRUS DRUGS VACCINATIONS (GENERIC) MASK, GOWN, PPE

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit based on ORBIS Intellectual Property (IP), CORDA.

9 A key word methodology based on patent abstract, claims and titles.

Are FP self-reported inventions interdisciplinary? FP self-reported inventions (patent families) are more interdisciplinary compared to a random subset of world patents. This can be a marker for higher market value10, but also – and maybe more importantly – higher societal value.

Primary and secondary WIPO classes are analysed as a proxy of interdisciplinary of invention (Ejermo, 2005). More than half of FP inventions (1,080 patent families) are classified in more than one WIPO technology class. The figure is twice as high as the average for the random sample of world patents (25%). This could imply that FP self-reported inventions (patent families) are more interdisciplinary than non FP inventions, or that the FP self-reported inventions are focused on more interdisciplinary technologies (i.e. WIPO classes). For example, Pharmaceutical, Biotechnology and Organic fine chemistry patent classes are strongly connected, and they also represent the highest share of FP self-reported inventions (patent families).

Figure 4 shows the average number of secondary technological classes covered by the invention (patent family), aggregated by WIPO technological class. As seen from the Figure, FP self-reported inventions cover more secondary technology classes than the worldwide random subset of inventions. The average FP self-reported invention covers 2.07 secondary classes, compared to the 1.67 secondary classes covered by the average random invention11.

Figure 4 Average number of secondary technology classes per inventions published in 2009-2018, by priority technology class 3,00 FP inventions SUBSET inventions

2,50

2,00

1,50

1,00

0,50

0,00 Micro-structural and nano-technology polymers chemistry, Macromolecular coating technology, Surface chemistry materials Basic metallurgy Materials, machines special Other games Furniture, engineering Chemical technology Audio-visual materials biological of Analysis for management methods IT technology Environmental Other consumer goods Organic fine chemistry Mechanical elements Control Semiconductors machines paper and Textile apparatus and processes Thermal energy apparatus, machinery, Electrical Measurement chemistry Food Telecommunications Pharmaceuticals Handling Transport technology Medical Biotechnology turbines pumps, Engines, technology Computer Machine tools Optics Civil engineering processes communication Basic Digital communication

Note: SUBSET inventions refer to the random sample of world main patents (inventions). Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit ORBIS Intellectual Property (IP), CORDA

10 (Miller, 2006) 11 This differences might be affected by different practices of patent offices. For example, if FP patents tend to be, on average, published more at the EPO than worldwide patents, and if these patent office has a habit of classifying patents in more technology classes, then this would partially explain the results. The data does not allow to analyse such an effect.

Figure 5 further visualizes the interdisciplinary of FP self-reported inventions. Close connections are observed in areas of biotechnology, pharmaceuticals and organic fine chemistry. Overall the majority of FP self-reported inventions are ‘interdisciplinary’ within a common sector such as chemistry. There are few unexpected cross-sector connections such as machine tools, semiconductors and surface technology, coating. The analysis paves the way towards a better understanding of how or to what extend FP self-reported patents bridge gaps between technology areas.

Figure 5 Network structure of links between technology classes of FP inventions published in

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit ORBIS Intellectual Property (IP), CORDA and own calculations. Notes: a) The graph shows the structure of interdisciplinary connections in FP self-reported inventions. Two kinds of connections are visualized: 1) maximum spanning tree, 2) N-1 strongest links (N = 35, the total number of fields). Size of nodes reflects the total number of inventions in a technology class. Width of connections reflects the total number of inventions in the two classes they connect (connections in each graph are normalized by the total size of nodes to make them comparable). b) Colours reflect sectors defined at http://www.wipo.int/ipstats/en/statistics/patents/pdf/wipo_ipc_technology.pdf

Where are FP self-reported inventions protected? FP inventions are protected by 61 different patent offices worldwide. The European markets are the main markets targeted for exploitation (75% of all FP inventions are protected by EPO, 74% in one of the EU28 Member State) followed by United States of America (74% of FP inventions are protected in the USA). 28% of the self-reported FP inventions are protected in China. On average each FP self-reported invention is protected in 3.7 different markets.

The choice of patenting office reflects the intention to use or license an invention for commercial application in a certain market. Inventions are more likely to be patented in large potential markets. The majority of FP self- reported patents are patented in Europe (75% at the European Patent Office, 74% in individual EU Member States). In Europe, apart from the EPO, FP inventions are mostly patented in the United Kingdom (19%), Spain, France, Poland and Germany (around 10% each). In addition to the home market, most FP inventions are protected in the United States of America (74%) and some on the Asian markets (28% of FP inventions are patented in China, 24% in Japan). 59% of the inventions are patented through the World Intellectual Property Organization (WIPO), an agency of the United Nations.

Figure 6 Share of FP self-reported inventions patented between 2009 and 2018, by patenting office

100%

90%

80% 75% 74% 74% 70% 59% 60%

50%

40% 28% 30% 24% 21% 20% 12% 10% 9% 10% 6%

0% EPO EU28 USA Canada Australia China Japan Korea India WIPO Others

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit ORBIS Intellectual Property (IP), CORDA. Note: a) FP patents published in 2009-2018, in all technology classes at the largest patent offices.

What is the market value of FP self-reported inventions? The estimated average market value of an FP patent is between EUR 72 000 and 334 000 depending on the market. In general FP patents are valued more than average patents in these markets with the exeption of the EPO and the USA. An average patent registered in the USA is valued higher than a patent registered in Europe, but the FP patents reach the highest valuation on the markets in Asia.

Patent market valuation is a method to compare the quality of inventions against each other in terms of its market attractiveness, technical quality, transferability to different industry etc. There are several such methods to determine patent values. ORBIS IP database, the database tested for the analysis of this Monitoring Flash, uses the method developed by IPBI12 which takes into account 26 indicators13 to estimate the market value of patents. Other valuation methods should be tested and compared in the future.

Based on IPBI estimates, the average value of FP self-reported patent is between EUR 72 000 and 334 000 depending on the market, higher than values of average patent in these markets. Patents registered in the US are, on average, valued much higher than those registered in Europe or Asia. The FP patents reach the highest valuation on the markets in Asia (i.e. in Korea, China and Japan).

Even if the analysis is descriptive in nature and cannot unpack the specific reasons behind the differences in estimated values, the figures reinforce the findings of the interim evaluation of Horizon 2020, that FP-funded research produces high-quality and valuable patents14

12 https://www.ip-bi.com/ The details of the evaluation methodology are not publicly available and were not assessed for the purpose of this Monitoring Flash. 13 Community application, R&D strength of the invention, R&D applicant ratio, Technology in different term trend, Sustainability of technology trend, Total size of activity, Family size, Transferability to different industries, Heterogeneity of potential applications, Exploitation in different technologies, Total amount of exploitation possibilities, Evidence of use, Relevance for other technologies/applications, Differentiation to the state of the art, Differentiation from direct competitor-technologies, Interfering with competitors technologies, Validity level, Patent maturity, Claim width and coverage, Validity in certain countries, Intended worldwide protection, Procedural State and Grant lag. 14 European Commission, 2017: p. 133

Figure 7 Average FP patent values compared to all patent values by major patenting office between 2009-2018 (IPBI estimated market value in thousands EUR)

400

350 334 Thousands 300

250 236 191 200 183 161 150 102 83 84 100 72 72 39 50 25

0 EPO + EU28 EPO USA China Japan Korea

FP Patent All Patents

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit ORBIS Intellectual Property (IP), CORDA. Notes: a) Average value of all the patents from patent families (invention) as calculated in the ORBIS Intellectual Property database.

The same is observed when analysing FP inventions by technology classes and comparing to worldwide averages. Figure 8 below shows that FP self-reported inventions are, on average, of higher value than the worldwide average in the majority of technology classes. Among the different classes, the biggest ‘FP Added Value’ (the additional FP patent value compared to the worldwide average) seems to be in the Handling (part of Mechanical engineering), Macromolecular chemistry fields and Food chemistry. However some FP patents are valued less than the average in particular in technologies related to Digital communication, Computer technology, Medical technology and Biotechnology.

Figure 8 Average market values of FP patents compared to all patents by technology class between 2009-2018 (IPBI estimated market value Thousands EUR)

1200 FP Patent All Patents

1000 Thousands

800

600

400

200

0 Macromolecular chemistry, polymers chemistry, Macromolecular Handling chemistry Food Micro-structural and nano-technology Pharmaceuticals materials biological of Analysis energy apparatus, machinery, Electrical Semiconductors games Furniture, metallurgy Materials, Telecommunications communication processes Basic chemistry materials Basic Control Optics Organic fine chemistry Transport Mechanical elements Digital communication Civil engineering for management methods IT Computer technology technology Medical Biotechnology Machine tools machines special Other turbines pumps, Engines, technology Environmental Measurement engineering Chemical apparatus and processes Thermal coating technology, Surface technology Audio-visual machines paper and Textile Other consumer goods

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit ORBIS Intellectual Property (IP), CORDA and own calculations. Notes: a) Average value of all the patents from patent families (invention) as calculated in the ORBIS Intellectual Property database

Framework Programme Patent Owners Who owns an FP invention? More than half (52%) of FP patent owners are small and medium-sized organisations (SMEs). This is much higher than the overall share of SME participation in the FP (around 20%), but much lower if compared to the World and the EU. 36% of FP patent owners are very large organisations. The majority of FP patent owners are located in Europe (75%).

A study by the EPO and EUIPO (Meniere et al., 2019) shows that companies that hold patents or other forms of intellectual property rights (IPR) are more likely to grow and experience high growth than those that do not. These indicates that the Framework Programmes can increase the competitiveness of the European economy by funding companies with high chances of innovating and patenting.

The owner of FP invention is the organisation that currently owns the self-reported patent in FP7 or Horizon 2020. When an organization is a subsidiary, the owner is the parent organization if holding the majority of shares. 52% of the FP patent owners are SMEs. The share is high if compared to the overall share of SME participation in FP, around 20%. However, the share is low if compared to the overall picture of patenting in Europe and the World where SMEs represent close to 80% of all patent owners. At the same time a high share of FP patent owners are very large organisations (36%) compared to the rest of Europe and the World (8% and 6%)15.

15 Organisations include commercial companies, but also universities and research centres for which information is available in ORBIS Europe dataset. The data contains at least 211 universities – based on their official names, of which 146 small and medium-sized, 9 large and 56 very large.

Figure 9 FP patent owners, by size of organisations

World EU28 FP Small Medium Large Very large

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit ORBIS Intellectual Property (IP), CORDA. Notes: Percentages are calculated based on 1,2 million entities for which data is available.

Further analysis shows that 31% of FP inventors are owned by organisations active in professional, scientific and technical sector, followed by manufacturing (29%), financial and insurance sector (11%) and education sector (11%) (Figure 11).

Figure 10 Share of FP invention owners by industry classification (main NACE rev. 2)

M - Professional, scientific and technical activities 31%

C - Manufacturing 29%

K - Financial and insurance activities 11%

P - Education 11%

G - Wholesale and retail trade; repair of motor vehicles and 4% motorcycles

N - Administrative and support service activities 3%

Q - Human health and social work activities 3%

J - Information and communication 2%

S - Other service activities 2%

Other 5%

Source: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit ORBIS Intellectual Property (IP), CORDA. Notes: a) The figure shows the percentage of applicants for the main patents from each FP patent family (invention). b) Calculations are based on 1549 entities for which industry and patent data was available. c) Other* relates to Electricity, gas, steam and air conditioning supply, Real estate activities, Construction, Human health and social work activities, Information and communication, Administrative and support service activities, Wholesale and retail trade; repair of motor vehicles and motorcycles, Water supply; sewerage; waste managment and remediation activities, Mining, Transporting and storage, Arts, entertainment and recreation, Agriculture, Accommodation and food service activities, Forestry, Activities of extraterritorial organisations and bodies

In terms of location, the majority of FP inventions are owned by organisations located in Europe (75%): 16% in Germany, 11% in Spain, 11% in the UK, 6% in France, 6% in Italy and 5% the Netherlands. 7% of the FP inventions are owned by organisations located in the US. 90% of the FP funding goes to organisations and individuals located in EU Member States.

Key messages This Monitoring Flash tested the possibilities of new analytical insights into how the Framework Programmes foster innovation through increased patenting activity. It aimed to do so by enhancing the internal EC data on patents with an external ORBIS Intellectual Property dataset. The key learning of this endeavour is that at present original FP patent data face significant quality and tracing challenges which limit the relevance of the analysis for wider policy. Data on FP patenting is rather patchy and a streamlined process of registering patents and other IP protection mechanisms and inventions stemming from FP-funded research is still needed. A large number of patents reported by beneficiaries were granted or applied for before the projects had started (background patents). In the future, the EC should increase its efforts and capacity to ensure appropriate and robust monitoring of IP protection mechanisms of FP beneficiaries for better informed policy.

Nevertheless, the exercise proved that a much more comprehensive analysis and understanding of the FP and its contribution to the patenting landscape can be conducted when EC data are enhanced with external company and patent databases. For instance, the analysis allows to conclude that FP self-reported inventions: . Mainly relate to health sector in areas such as biotechnology, pharmaceuticals or organic chemistry; . Are more interdisciplinary than random patents; . Have in general a higher estimated values compared to the market averages; . Are intended to be largely exploited in Europe and the United States; . 31% are owend by organisations active in professional, scientific and technical sector; . The majority of FP inventions are owned by organisations located in Europe (75%). . For the future R&I policy this analysis: . Confirms a considerable time-lag between FP activities and potential exploitation which should be taken into account with the future Horizon Europe programming focused on the targeted impacts. Attention should be given to the results of the past programmes; . Identifies a need of a much deeper analysis to better understand and monitor the existing patent and other IPR classifications against the EU policy objectives in particular related to climate objectives; . Highlights the value that evidence can bring to design better policies in the future.

References Bronzini, R. and Piselli, P. (2016). The impact of R&D subsidies on firm innovation. Research Policy, 45(2):442–457. Cohen W., Nelson R., Walsh J. (2000). Protecting their intellectual assets: appropriability conditions and why US manufacturing firms patent (or not). National Bureau of Economic Research Working Paper No. 7552, Cambridge, MA. Ejermo, O. (2005). Technological diversity and Jacobs’ externality hypothesis revisited. Growth and Change, 36(2), 167- 195. European Commission (2017). Interim Evaluation of Horizon 2020. Staff Working Document. SWD(2017)220. European Commission (2018). Impact Assessment of Horizon Europe. Staff Working Document. SWD(2018)307. Hall B., Helmers C., Rogers M., Sena V. (2013). The importance (or not) of patents to UK firms. Oxford Economic Papers , 65 , 603 – 29. Howell, S. T. (2017). Financing Innovation: Evidence from R&D Grants. American Economic Review, 107(4):1136–1164. Meniere, M., Rudyk, Y., Wajsman, I., Kazimierczak, N. (2019). High-growth firms and intellectual property rights. IPR profile of high-potential SMEs in Europe. EPO & EUIPO Report May 2019. ISBN 978-3-89605-228-5. Miller, D. J. (2006). Technological diversity, related diversification, and firm performance. Strategic Management Journal, 27(7), 601-619. Pénin, J. & Neicu, D. (2018). Patents and Open Innovation: Bad Fences Do Not Make Good Neighbors. Journal of Innovation Economics & Management, 25(1), 57-85. Pisano, G. P., Teece, D. (2007), How to Capture Value from Innovation: Shaping Intellectual Property and Industry Architecture, California Management Review, 50(1), 278-295. Pries, F., Guild, P. (2011), Commercializing Inventions Resulting from University Research: Analyzing the Impact of Technology Characteristics on Subsequent Business Models, Technovation, 31(4), 151-160. Wildmann, R. (2017). The effect of government research grants on firm innovation: theory and evidence from Austria.

ANNEX Methodological Note Datasets

The main dataset used for our analysis originates from CORDA, the European Commission’s common research data

warehouse. CORDA is the Framework Programmes' central repository of data collected and derived during the course of FP © European Union, 2020 implementation.

Data obtained from CORDA includes patent publication numbers, which allow linking to the ORBIS Intellectual Property database.16 Following this link, information on the valuation of patents, application and publication dates, patent offices, WIPO technology classes and IPC classifications, as well as identification numbers of the patents’ applicants and owners were collected and analysed. Information on patent applicants and owners was retrieved from ORBIS Europe.17

FP patents

The main patent dataset containes 4,616 patent families (20,821 individual patents) reported by FP7 and H2020 projects, extended to include all their family members by linking with ORBIS Intellectual Property. Only EC-verified patents are included. The verification process includes several steps, after which 68% of reported patents are included in the analysis dataset: 90% of reported patents were checked, out of which 81% were validated. Finally, 92% of the validated patents were found in ORBIS Intellectual Property.

In the last step, only foreground patents were selected for present analysis. Foreground patents are those with application or first priority date later than one year after the start of the first project in which a patent was reported. 2 776 foreground patent families (inventions), comprising 10 920, foreground patents we used for the analysis.

Data on the owners of FP patents (FP inventors) was obtained from ORBIS Intellectual Property and ORBIS Europe. The databases offers information on 1 860 current FP patent owners. Depending on the information analysed, the samples can be smaller by 10%.

Random sample of worldwide patents

To compare FP-reported patents to the worldwide IPR environment, we have proceeded in two ways. First, where worldwide data is available from ORBIS Intellectual Property, we compare FP patents to the aggregated figures. For sections where such data is not available, we compare FP patents with a random sample of 11,998 patent families (48,131 patents) selected from ORBIS Intellectual Property and matching the distribution of FP patents by publication year.18

Other monitoring flash reports available here: https://ec.europa.eu/info/publications/horizon-2020-monitoring-flash_en

#1 Country Participation #2 Dynamic Network Analysis #3 International Cooperation #5 Sustainable Development Goals

16 https://orbisintellectualproperty.bvdinfo.com 17 https://orbiseurope.bvdinfo.com 18 The random patent sample was constructed in four steps i)random subset of 500,528 patents from the Google Patents database was downloaded (publication numbers and publication dates); ii) out of these patents, a random subset of 20,000 patents was selected, such that the probability of selection was higher for years with higher number of FP patents; iii) this dataset was then extended by all the family members from ORBIS Intellectual Property; iv) finally, a random subset of 12,000 families (inventions) was drawn to match the publication year distribution of foreground FP patents.

DOP O

by Martina Kadunc

FROM HORIZON 2020 TO HORIZON EUROPE MONITORING FLASH SUSTAINABLE DEVELOPMENT GOALS

February 2020

This Monitoring Flash elaborates on how Horizon 2020 – the European Framework Programme for Research and Innovation 2014-2020 – contributes to the achievement of the Sustainable Development Goals (SDGs). The SDGs are a universally agreed blueprint1 adopted by the United Nations in 2015 for achieving a better and more sustainable future for all. The EU has committed to fully implement the SDG agenda and the underlying social and economic transitions it requires.2 EU Research and Innovation (R&I) investment is one of the main drivers and enablers of this transition. This requires, however, a much better overview of the R&I efforts and their added value for the SDGs. This Monitoring Flash provides a starting point for discussion, acknowledging the overall complexity related to data and methodologies.

HORIZON 2020 FOR SDGs Key overview data

€37.7b 20,994 29,086 3 of Horizon 2020 projects relevant for SDGs participants working on SDGs targeted by a investment relevant for (83%) projects relevant for SDGs Horizon 2020 project on SDGs (84%) (91%) average

MAJORITY OF THE HORIZON 2020 INVESTMENT EXPECTED TO FOSTER SDGs Key overall messages . Potentially up to 84% of the Horizon 2020 investment relates to at least one of the SDGs and all pillars of Horizon 2020 appear to contribute to the SDGs to a similar degree (‘top-down’ and ‘bottom-up’ investment). . The largest share of the investment relates to Climate action (SDG 13) and Good health and well-being (SDG 3). The focus of the EU R&I investment on Responsible production and Consumption is low especially when compared to the current EU performance gap to SDG targets in this area. . A typical project in Horizon 2020 relates to three different SDGs indicating that the investments are highly interconnected across the SDG spectrum. . Future policy choices should consider the systemic nature and interdependence of SDGs. Key opportunities lay in leveraging the synergies and minimising the potential trade-offs among the different SDGs (e.g. in Horizon 2020 some interlinkages such as climate, energy and water are expected, and others such as climate action and good health are more surprising).

1 https://sustainabledevelopment.un.org/?menu=1300 2 https://www.consilium.europa.eu/en/press/press-releases/2017/06/20/agenda-sustainable-development/

Introduction

The UN Sustainable Development Goals (SDGs) present a global political agenda addressing a range of today’s most pressing social, economic, and environmental challenges3. The EU has committed to implement the SDGs across all policies, including in Research and Innovation (R&I), and modernise the European economy and society to achieve a sustainable future. In her political guidelines, President von der Leyen made the achievement of the SDGs the responsibility of governments and the European Institutions.

The EU Framework Programme for R&I is one of the main vehicles to direct the future transitions by investing in new knowledge and solutions that will enable Europe to become the first climate-neutral continent in the world by 2050, to grasp the opportunities of the digital age and to accelerate Europe’s recovery, preparedness and resilience. As such the R&I investment forms an integral part of an ambitious recovery plan and a modernised 7- year budget for the EU. The recovery will also capitalise on the results of Horizon 2020 and its predecessors.

Within the SDG landscape, all R&I investments by design contribute towards building a resilient infrastructure, promoting inclusive and sustainable industrialisation and fostering innovation (SDG 09). In addition, some R&I investments pursue specific societal objectives such as good health and wellbeing or tackling climate change. This Monitoring Flash aims to better understand the latter – the societal orientation of Horizon 2020.

The Monitoring Flash methodology is based on a keyword search. For each of the SDGs, a list of keywords was assembled based on a compilation and cross-checking of keywords used for similar endeavours (i.e. Mapping Austrian Research Contributions to the Sustainable Development Goals, Aurora Universities Network SDG analysis and Australian Guide for Getting Started with the SDGs in Universities). The keyword search (included in the Annex) was applied using IRIS, a Commission internal search and discovery tool which allows text mining of 850 000 unstructured text documents related to Horizon 2020 proposals and project deliverables4. Portfolios of documents were linked with the Horizon 2020 Dashboard to obtain the final dataset for analysis based on Horizon 2020 monitoring data. The advantages of this method are: i) a bottom-up analysis based on individual project proposals and deliverables; ii) possibility for regular, automatic updates of the analysis and iii) ability to apply key words to different sets of documents (e.g. publications, patents, policy documents). The disadvantages of this method are: i) it cannot distinguish between a project mentioning some SDG related concept and a project aiming to work or contribute towards a certain SDG in a more substantial manner and ii) the precision of methodology can only increase with a substantial and regular expert input. With a growing importance of the SDG agenda for R&I, there is also an increasing risk of ‘SDG-washing’ (ie. when information is used to give the impression that a project is SDG relevant even when in practice, this is not the case).

The present key word methodology is novel for the Framework Programme monitoring and applied for the first time in this Monitoring Flash. As such the figures do not represent official statistics but rather show a potential for AI driven monitoring in the future. In particular, in relation with the 35% Horizon 2020 climate action target, the official monitoring based on a more conservative RIO marker methodology should be considered.

Against this caveats, the Monitoring Flash - for the first time - estimates the Horizon 2020 SDG landscape. The analysis aims at launching a discussion on the future EU R&I policy and the necessary advancement in data and methodologies needed for improved policy intelligence.

3 In September 2015, 193 Member States of the United Nations adopted 17 Sustainable Development Goals. 4 Relevance rankings were used to obtain the final portfolios of documents for analysis.

The majority of Horizon 2020 investment expected to foster SDGs

Potentially up to 84% of the current Horizon 2020 investments relate to at least one SDG. Overall this represents EUR 37.7 billion invested in 20,994 different projects, carried out by 29,086 different beneficiaries from 152 countries5. For comparison, the ex-post evaluation of FP7 estimated that 72% of the FP7 ‘Cooperation’ and ‘Capacities’ programmes 6 investment could have positive expected impacts on SDGs. It is estimated that the largest share of the present Horizon 2020 portfolio relates to Goals of Climate Action and Good Health and Well-Being. A comparatively small share of the investments relate to No Poverty, Fostering Gender Equality and Global Partnerships for Sustainable Development. 16% of investment did not relate to any of the SDGs and can be understood as non-oriented R&I such as mathematics and physics for instance.

All the SDGs are equally important for the sustainable future and there are many potential sustainability pathways. When it comes to prioritising the EU R&I investment towards the achievement of the SDGs, importance may be given to the pathways where leaps in knowledge, technology or innovation can make the biggest difference. Another option is to focus on areas where the ‘EU distance to SDG targets’ remains large. Based on the 2019 Sustainable Development Report7 the EU’s largest gap towards achieving the 2030 Goals relates to Responsible Consumption and Production (EU is almost 20 percentage points away from reaching the targets)8 followed by Climate Action (EU is around 10 percentage points away from reaching the targets under SDG 13). However, the EU has almost reached the Goals and associated targets on No Poverty, Good Health and Well-Being and Affordable and Clean Energy.

Figure 1 Share of potential Horizon 2020 investment by SDG

EU distance to SDG target (in percentage points from 0.3 p.p. SDG 1 to 17.3 p.p. SDG 12)

Sources DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, IRIS, Horizon 2020 Dashboard, CORDA Cut-off date 1 July 2019 and the Sustainable Development Report 2019

5 Based on RIO Markers Methodology the Horizon 2020 investment in climate action, sustainable development and biodiversity is at 62%. The RIO Markers distinguish between principal objective, secondary objective or not targeted at all corresponding to the values 100%/40%/0% (the 'markers'). Markers are allocated manually: for programmable actions, markers are allocated at topic level and automatically matched with projects selected for funding. For bottom-up actions the markers are allocated once projects are selected for funding. RIO Markers Methodology is used to monitor Horizon 2020 climate expenditure target aimed at 35% (based on RIO Markers the current expenditure is at 30%). 6 FP7-COOPERATION programme represented 64% (EUR 28 million) of the FP7 and aimed at stimulating EU wide collaborative research. The methodology applied for SDG assessment was different, based on expert review. FP7 and the SDGs 7 Bertelsmann Stiftung and Sustainable Development Solutions Network, Sustainable Development Report, June 2019 Figure 19. The European Commission Joint Research Centre (JRC) conducted an independent statistical audit of the report’s methodology and results. 8 The Sustainable Development Report 2019 describes countries’ progress towards achieving the SDGs and indicate areas requiring faster progress. The SDG Index score and scores by goal can be interpreted as a percentage of achievement. The difference between 100 and countries’ scores is therefore the distance in percentage that needs to be completed to achieving the SDGs and goals.

The focus of the EU R&I investment in responsible production and consumption is low when compared to the EU performance gap to achieve the SDG targets in this area. 28% of current Horizon 2020 investment (EUR 12.5 billion) potentially relates to Goal 12. The ex-post evaluation of FP7 ‘Cooperation’ and ‘Capacities’ programme estimated that the share of projects related to Goal 12 was at a similar level of 25% of the total investment9.

All the three pillars of the programme potentially contribute to the SDGs to a similar degree. A high relevance to the SDGs is expected from the ‘top-down’ policy focused Pillar 3 aimed at tackling societal challenges (82% of the current investments under this pillar could relate to SDGs). However, it is interesting that the ‘bottom-up’ investigator-driven pillars – such as Pillar 1 Excellence Science which includes the European Research Council, Marie Skłodowska-Curie Actions, and Research Infrastructures – also seem to foster a high degree of SDGs relevant excellent science and high-quality knowledge (86% of the current investments under this pillar could relate to SDGs).

Figure 2 Share of Horizon 2020 investment relevant to SDGs, by pillar

Sources: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, IRIS, Horizon 2020 Dashboard, CORDA Cut-off date 1 July 2019 Note: RIO Markers is a methodology applied to monitor Horizon 2020 expenditure in climate action, sustainable development and biodiversity. It distinguishes between principal objective, secondary objective or not targeted and accounts the related share of investment as 100%/40%/0% (the 'markers'). Markers are allocated manually: for programmable actions (e.g. Pillar 3) markers are allocated at topic level and automatically matched with projects selected for funding. For bottom-up actions (e.g. Pillar 1) the markers are allocated once projects are selected for funding. RIO Markers Methodology is by design a more conservative methodology compared to the key word methodology tested in this Flash.

SDGs are addressed by all the policy instruments of Horizon 2020 to a similar degree. SDG relevant projects are observed in all the different types of actions in Horizon 2020 from collaborative R&I projects to individual research grants and support for innovators as well as other instruments such as public procurement and coordination and support actions.

Whereas the present assessment of SDG related Horizon 2020 investment is encouraging, there is a need for a substantial and continuous effort to better understand the changing dynamics of R&I. The Sustainable Development Agenda can serve as a compass to increase the relevance and benefits of R&I for citizens.

9 FP7 and the SDGs

High degree of interlinkages among the SDGs

Achieving systemic change is complex and requires a move away from conventional silos, sectors and disciplines towards a more interconnected and systemic thinking. The SDG Agenda explicitly recognises the interlinkages among the different Goals and the existence of both the trade-offs and co-benefits amongst them. The key opportunity lies in making sure that the policy choices leverage the co-benefits and alleviate the potential trade-offs.

Based on the latest Global Sustainable Development Repot 201910 only about 10% of possible SDG interactions have been studied so far and knowledge gaps remain. Some SDGs bring co-benefits across the Goals namely SDG 3 Good Health and Well-Being, SDG 4 Quality Education, SDG 5 Gender Equality and SDG 17 Partnerships for the Goals. Other SDGs come with more significant trade-offs including SDG 1 No Poverty, SDG 2 Zero Hunger, SDG 7 Affordable and Clean Energy and SDG 8 Decent Work and Economic Growth.

The Horizon 2020 mapping against the SDGs shows that the current investment is indeed highly interconnected (Figure 3). An average Horizon 2020 project potentially relates to three different Goals. Whereas the current methodology is not fit to identify whether the projects foster synergies or trade-offs among the individual Goals, some useful policy insights emerge. For instance, the climate action Goal seems to be strongly connected not only to energy, cities and clean water but also to health related Goals. The energy Goal is highly connected to cities and climate action.

Figure 3 Overview of the interlink between SDG-related Horizon 2020 investment, by SDG

Sources: DG Research and Innovation, Programme Analysis & Regulatory Reform Unit, IRIS, Horizon 2020 Dashboard, CORDA Cut-off date 1 July 2019 and visualisation by Flouris

10 See Box 1-2 Assessing interaction among Sustainable Development Goals

SDG Monitoring and Policy Outlook for R&I

This Monitoring Flash tested the possibilities of portfolio mapping of Horizon 2020 investment to date against the Sustainable Development Goals. It aimed to do so by using IRIS, the Commission’s latest internal search and discovery tool for Horizon 2020 projects. Whereas the analysis results in a general overview of the Horizon 2020 SDG portfolio, any assessment to inform policy priorities needs to be further refined with considerable expert input and additional cross-categorisation by technology areas, research fields etc. In addition, more research is needed to first better understand where leaps in knowledge, technology or innovation can make the biggest difference and where the R&I policy choices can leverage the co-benefits and hamper the potential trade-offs amongst the SDGs.

As for the SDG monitoring in the future Horizon Europe programme, the considerable expert input needed for better policy insight could be dispersed by systemic ex-ante integration of the SDGs in project proposals, evaluations and reporting of results. Furthermore individual ‘SDG champions’ could be tasked to develop in- depth knowledge of specific SDGs in the Framework Programme and the main R&I entry points within them.

UP TO 12% OF HORIZON 2020 INVESTMENT (EUR 5.4 BILLION) POTENTIALLY RELATES TO ENDING POVERTY IN ALL ITS FORMS EVERYWHERE.

SDG 1 is one of the least addressed Goals by the EU R&I Framework Programme. The highest share of potential No Poverty related investment is identified in the ‘top-down’ parts of Horizon 2020, the Societal Challenges (up to 21% of Horizon 2020 investment in this pillar potentially relates to SDG 1). Significantly lower share of investment is observed in the bottom-up pillars (Excellent Science and Industrial Leadership). Based on the most frequent key words analysis, the projects seem diverse. A group of them mentions energy poverty and disaster risk reduction. Four out of five projects are also relevant for Climate Action (SDG 13), Sustainable Cities and Communities (SDG 11) and Reduced Inequalities (SDG 10). Based on the latest Global Sustainable Development Repot 201911 the main known trade-offs lie with the SDG 06 on Clean Water and Sanitation and SDG 15 Life on Land.

TOP 10 MOST ACTIVE ORGANISATIONS % PROJECTS & INVESTMENT % PROJECTS INTERLINKING WITH other SDGs

FRAUNHOFER GESELLSCHAFT ZUR 1 FOERDERUNG DER ANGEWANDTEN FORSCHUNG E.V. CENTRE NATIONAL DE LA RECHERCHE 2 SCIENTIFIQUE CNRS 3 CONSIGLIO NAZIONALE DELLE RICERCHE 4 UNIVERSITY COLLEGE LONDON ETHNIKO KENTRO EREVNAS KAI 5 TECHNOLOGIKIS ANAPTYXIS FUNDACION TECNALIA RESEARCH & 6 INNOVATION ALMA MATER STUDIORUM - UNIVERSITA 7 DI BOLOGNA 8 KATHOLIEKE UNIVERSITEIT LEUVEN 9 KOBENHAVNS UNIVERSITET THE CHANCELLOR, MASTERS AND 10 SCHOLARS OF THE UNIVERSITY OF OXFORD

WORD CLOUD - MOST SIGNIFICANT KEY WORDS USED IN HORIZON 2020 PROJECTS RELATED TO SDG 1

ADDRESSING SDG 1 IN HORIZON 2020: SAVING ON HEATING ENERGY AirEx is a smart ventilation control developed by a UK-based start-up focusing on the development and rapid scale-up of fabric efficiency technologies in homes to tackle energy poverty. Their solution enables significant savings on heating energy bills for residents, without compromising air quality.

11 See Box 1-2 Assessing interaction among Sustainable Development Goals

UP TO 17% OF HORIZON 2020 INVESTMENT (EUR 7.7 BILLION) POTENTIALLY RELATES TO ENDING HUNGER, ACHIEVING FOOD SECURITY, IMPROVED NUTRITION AND PROMOTING SUSTAINABLE AGRICULTURES.

SDG 2 Zero Hunger is moderately addressed by the EU R&I Framework Programme. The highest share of potential zero hunger related investment is identified in the ‘top-down’ parts of Horizon 2020, the Societal Challenges (up to 27% of Horizon 2020 investment in this pillar potentially relates to SDG 2). Based on the most frequent key words analysis, projects seem to focus on nutrition security and food systems. Four out of five projects are also relevant to Climate Action (SDG 13). Based on the latest Global Sustainable Development Repot 201912 the main currently known trade-offs lie with the SDG 11 Sustainable Cities and Communities, SDG 14 Life Below Water, SDG 15 Life On Land and SDG 6 Clean Water and Sanitation – every other Zero Hunger related project in Horizon 2020 potentially tackles these interlinkages.

TOP 10 MOST ACTIVE ORGANISATIONS % PROJECTS & INVESTMENT % PROJECTS INTERLINKING WITH other SDGs

INSTITUT NATIONAL DE LA RECHERCHE 1 AGRONOMIQUE CENTRE NATIONAL DE LA RECHERCHE 2 SCIENTIFIQUE CNRS 3 STICHTING WAGENINGEN RESEARCH 4 CONSIGLIO NAZIONALE DELLE RICERCHE AGENCIA ESTATAL CONSEJO SUPERIOR 5 DEINVESTIGACIONES CIENTIFICAS FRAUNHOFER GESELLSCHAFT ZUR FOERDERUNG DER ANGEWANDTEN 6 FORSCHUNG E.V. 7 WAGENINGEN UNIVERSITY 8 KOBENHAVNS UNIVERSITET 9 AARHUS UNIVERSITET 10 DANMARKS TEKNISKE UNIVERSITET

WORD CLOUD - MOST SIGNIFICANT KEY WORDS USED IN HORIZON 2020 PROJECTS RELATED TO SDG 2

ADDRESSING SDG 2 IN HORIZON 2020: NEW ROBOTS TRANFROMING FARMING Flourish developed adaptable ground and aerial robots that can increase agricultural yield, reduce pesticide use and mitigate food security risks.

12 See Box 1-2 Assessing interaction among Sustainable Development Goals

UP TO 53% OF HORIZON 2020 INVESTMENT (EUR 23.8 BILLION) POTENTIALLY RELATES TO ENSURING HEALTH LIVES AND PROMOTING WELL-BEING FOR ALL AT ALL AGES.

SDG 3 Good Health and Well-Being is one of the most well addressed Goals by the EU R&I Framework Programme. Almost one out of two Horizon 2020 projects is potentially related to Good health and well-being, both in top-down and bottom-up parts of the programme. Based on the most frequent key words analysis, projects seem to focus on middle and low income countries and infectious diseases in particular the Zika virus. Every second Good health and well-being project is potentially related also to Climate Action goals (SDG 13). Based on the latest Global Sustainable Development Repot 201913 there are almost no currently known trade-offs when tackling this Goal apart from the Affordable and Clean Energy (SDG 7). However, major co-benefits are observed with SDG 8 Decent Work and Economic Growth.

TOP 10 MOST ACTIVE ORGANISATIONS % PROJECTS & INVESTMENT % PROJECTS INTERLINKING WITH other SDGs

CENTRE NATIONAL DE LA 1 RECHERCHE SCIENTIFIQUE CNRS FRAUNHOFER GESELLSCHAFT ZUR 2 FOERDERUNG DER ANGEWANDTEN FORSCHUNG E.V. 3 UNIVERSITY COLLEGE LONDON AGENCIA ESTATAL CONSEJO 4 SUPERIOR DEINVESTIGACIONES CIENTIFICAS CONSIGLIO NAZIONALE DELLE 5 RICERCHE THE CHANCELLOR, MASTERS AND 6 SCHOLARS OF THE UNIVERSITY OF OXFORD 7 KOBENHAVNS UNIVERSITET THE CHANCELLOR MASTERS AND 8 SCHOLARS OF THE UNIVERSITY OF CAMBRIDGE MAX-PLANCK-GESELLSCHAFT ZUR 9 FORDERUNG DER WISSENSCHAFTEN EV KATHOLIEKE UNIVERSITEIT 10 LEUVEN

WORD CLOUD - MOST SIGNIFICANT KEY WORDS USED IN HORIZON 2020 PROJECTS RELATED TO SDG 3

ADDRESSING SDG 3 IN HORIZON 2020: RESEARCH TO SHIELD AGAINST ZIKA-RELATED BIRTH DEFECTS ZIKAlliance project research found out that dengue virus infection in fact offers protection against the development of Zika- related birth defects rather than enhance its development.

13 See Box 1-2 Assessing interaction among Sustainable Development Goals

UP TO 20% OF HORIZON 2020 INVESTMENT (EUR 9 BILLION) POTENTIALLY RELATES TO ENSURING INCLUSIVE AND EQUITABLE QUALITY EDUCATIONS AND PROMOTING LIFELONG LEARNING OPPORTUNITIES FOR ALL.

SDG 4 Quality Education is moderately addressed by the EU R&I Framework Programme. The highest share of potential Quality Education related investment is identified in the top-down parts of the Horizon 2020 Programme related to Societal Challenges and Other priorities such as widening and including society. Based on the most frequent key words analysis projects seem to focus on early childhood education, science education and teachers. More than half of the projects interlink with the SDG 3 on Good Health and Well-Being. Based on the latest Global Sustainable Development Repot 201914 tackling Quality Education targets have no currently known trade-offs with other SDGs and co-benefits are observed across all the other Goals.

TOP 10 MOST ACTIVE ORGANISATIONS % PROJECTS & INVESTMENT % PROJECTS INTERLINKING WITH other SDGs

CENTRE NATIONAL DE LA RECHERCHE 1 SCIENTIFIQUE CNRS FRAUNHOFER GESELLSCHAFT ZUR 2 FOERDERUNG DER ANGEWANDTEN FORSCHUNG E.V. 3 UNIVERSITY COLLEGE LONDON 4 KATHOLIEKE UNIVERSITEIT LEUVEN 5 KOBENHAVNS UNIVERSITET 6 CONSIGLIO NAZIONALE DELLE RICERCHE THE CHANCELLOR, MASTERS AND 7 SCHOLARS OF THE UNIVERSITY OF OXFORD AGENCIA ESTATAL CONSEJO SUPERIOR 8 DEINVESTIGACIONES CIENTIFICAS ETHNIKO KENTRO EREVNAS KAI 9 TECHNOLOGIKIS ANAPTYXIS THE CHANCELLOR MASTERS AND 10 SCHOLARS OF THE UNIVERSITY OF CAMBRIDGE

WORD CLOUD - MOST SIGNIFICANT KEY WORDS USED IN HORIZON 2020 PROJECTS RELATED TO SDG 4

ADDRESSING SDG 4 IN HORIZON 2020: GAMES PLATFORM TO MAKE LEARNING FUN The BEACONING worked with schools and universities from Bulgaria, France, Greece, Italy, Poland, Romania, the United Kingdom, Turkey, Israel and South Africa, to create an innovative games platform to make learning science, technology, engineering and maths (STEM) fun. Their platform lets teachers pick a narrative and choose from a series of mini games.

14 See Box 1-2 Assessing interaction among Sustainable Development Goals

UP TO 5% OF HORIZON 2020 INVESTMENT (EUR 2.1 BILLION) POTENTIALLY RELATES THOWARDS ACHIEVING GENDER EQUALITY AND EMPOWERMENT OF ALL WOMEN AND GIRLS.

SDG 5 Gender Equality Goal is one of the least addressed Goals by the EU R&I Framework Programme. The highest share of gender equality related investment is identified in the ‘other’ parts of the Horizon 2020 including society and widening (up to 12% of Horizon 2020 investment in this programme part potentially relates to SDG 5). Based on the most frequent key words analysis projects seem to focus on feminism and gender violence. Every three out of five gender equality related projects also interlink with Peace, Justice and Strong Institutions. Based on the latest Global Sustainable Development Repot 201915 tackling Gender Equality targets has almost no currently known trade-offs with other SDGs and co-benefits have been observed across all the other Goals.

TOP 10 MOST ACTIVE ORGANISATIONS % PROJECTS & INVESTMENT % PROJECTS INTERLINKING WITH other SDGs

CENTRE NATIONAL DE LA RECHERCHE 1 SCIENTIFIQUE CNRS FRAUNHOFER GESELLSCHAFT ZUR 2 FOERDERUNG DER ANGEWANDTEN FORSCHUNG E.V. 3 UNIVERSITY COLLEGE LONDON ETHNIKO KENTRO EREVNAS KAI 4 TECHNOLOGIKIS ANAPTYXIS 5 UNIVERSITEIT UTRECHT 6 UNIVERSITEIT VAN AMSTERDAM 7 KATHOLIEKE UNIVERSITEIT LEUVEN THE CHANCELLOR, MASTERS AND 8 SCHOLARS OF THE UNIVERSITY OF OXFORD 9 KENTRO MELETON ASFALEIAS THE CHANCELLOR MASTERS AND 10 SCHOLARS OF THE UNIVERSITY OF CAMBRIDGE

WORD CLOUD - MOST SIGNIFICANT KEY WORDS USED IN HORIZON 2020 PROJECTS RELATED TO SDG 5

ADRESSING THE SDG 5 IN HORIZON 2020: THE GENDER EFFECTS OF CHANGES IN TAX BASES The Swedish FairTax project carried out in-depth comparative, interdisciplinary research using constitutional, legal, technical, institutional, qualitative, and quantitative methods to address more sustainable and fair tax and social policies including microsimulation results for personal income tax systems in six countries of the European Union.

15 See Box 1-2 Assessing interaction among Sustainable Development Goals

UP TO 38% OF HORIZON 2020 INVESTMENT (EUR 17.1 BILLION) POTENTIALLY RELATES TO ENSURING AVAILABILITY AND SUSTAINABLE MANAGEMENT OF WATER AND SANITATION FOR ALL.

SDG 6 Clean Water and Sanitation is moderately addressed by the EU R&I Framework Programme. The highest share of clean water and sanitation projects is identified in the ‘top-down’ and ‘other’ parts of the Horizon 2020 Programme. Based on the most frequent key words analysis, projects relate to wastewater treatment and water reuse. Every four out of five projects seem to interlink with the Climate Action (SDG 13). Linkages with Good Health and Well-Being Goal are also frequent. Clean Water and Sanitation Goal is one of the most commonly assessed Goals for potential trade-offs and co- benefits. Based on the latest Global Sustainable Development Repot 201916 trade-offs were found with the Zero Hunger Goal (SDG 2) and Affordable and Clean Energy (SDG 07). Co-benefits are observed with all the other SDGs.

TOP 10 MOST ACTIVE ORGANISATIONS % PROJECTS & INVESTMENT % PROJECTS INTERLINKING WITH other SDGs

FRAUNHOFER GESELLSCHAFT ZUR 1 FOERDERUNG DER ANGEWANDTEN FORSCHUNG E.V. CENTRE NATIONAL DE LA 2 RECHERCHE SCIENTIFIQUE CNRS CONSIGLIO NAZIONALE DELLE 3 RICERCHE AGENCIA ESTATAL CONSEJO 4 SUPERIOR DEINVESTIGACIONES CIENTIFICAS COMMISSARIAT A L ENERGIE 5 ATOMIQUE ET AUX ENERGIES ALTERNATIVES KATHOLIEKE UNIVERSITEIT 6 LEUVEN FUNDACION TECNALIA RESEARCH 7 & INNOVATION TEKNOLOGIAN TUTKIMUSKESKUS 8 VTT OY DANMARKS TEKNISKE 9 UNIVERSITET UNITED KINGDOM RESEARCH AND 10 INNOVATION

WORD CLOUD - MOST SIGNIFICANT KEY WORDS USED IN HORIZON 2020 PROJECTS RELATED TO SDG 6

ADRESSING THE SDG 6 IN HORIZON 2020: SCALABLE SOLUTION FOR CLEANING WASTE WATER iMETland is a solution to clean polluted waste water being low-cost, eco-friendly, and scalable to an individual, a municipal, or an industrial use setting. The Spanish project is currently being commercialized by iMETfilter SL.

16 See Box 1-2 Assessing interaction among Sustainable Development Goals

UP TO 41% OF HORIZON 2020 INVESTMENT (EUR 18.6 BILLION) POTENTIALLY RELATES TO ENSURING ACCESS TO AFFORDABLE, RELIABLE, SUSTAINABLE AND MODERN ENERGY FOR ALL.

SDG 7 Affordable and Clean Energy is one of the most well addressed Goals by the EU R&I Framework Programme. The highest share of affordable and clean energy related projects is identified in Pillar 2 Industrial Leadership (up to 60% of Horizon 2020 investment in this pillar potentially relates to SDG 7). Based on the most frequent key words analysis, projects often refer to renewable energy systems and storage solutions. Every four out of five projects interlink with Climate Action (SDG 13). Links with the Sustainable Cities and Communities are also frequent. Based on the latest Global Sustainable Development Repot 201917 affordable and clean energy has many trade-offs across the SDGs namely SDG 6 Clean Water and Sanitation, SDG 2 Zero Hunger, SDG 10 Reduced Inequalities and SDG 15 Life on Land. Co-benefits were identified with SDG 12 Responsible Consumption and Production, SDG 4 Quality Education and SDG 5 Gender Equality.

TOP 10 MOST ACTIVE ORGANISATIONS % PROJECTS & INVESTMENT % PROJECTS INTERLINKING WITH other SDGs

FRAUNHOFER GESELLSCHAFT ZUR 1 FOERDERUNG DER ANGEWANDTEN FORSCHUNG E.V. CENTRE NATIONAL DE LA 2 RECHERCHE SCIENTIFIQUE CNRS COMMISSARIAT A L ENERGIE 3 ATOMIQUE ET AUX ENERGIES ALTERNATIVES CONSIGLIO NAZIONALE DELLE 4 RICERCHE TEKNOLOGIAN TUTKIMUSKESKUS 5 VTT OY FUNDACION TECNALIA RESEARCH & 6 INNOVATION 7 TECHNISCHE UNIVERSITEIT DELFT 8 DANMARKS TEKNISKE UNIVERSITET AGENCIA ESTATAL CONSEJO 9 SUPERIOR DEINVESTIGACIONES CIENTIFICAS NEDERLANDSE ORGANISATIE VOOR TOEGEPAST 10 NATUURWETENSCHAPPELIJK ONDERZOEK TNO

WORD CLOUD - MOST SIGNIFICANT KEY WORDS USED IN HORIZON 2020 PROJECTS RELATED TO SDG 7

ADRESSING THE SDG 7 IN HORIZON 2020: FACILITATING SYNERGIES FOR SUSTAINABLE CONSTRUCTION STORM project developed an innovative district heating and cooling network controller, based on self-learning algorithms, deployed and tested in two demo sites. The STORM controller reduces peaks in district heating networks and interacts with the electricity market reducing operational and investment costs for network operators.

17 See Box 1-2 Assessing interaction among Sustainable Development Goals

UP TO 19% OF HORIZON 2020 INVESTMENT (EUR 8.7 BILLION) POTENTIALLY RELATES TO PROMOTING SUSTAINED, INCLUSIVE AND SUSTAINABLE ECONOMIC GROWTH, FULL AND PRODUCTIVE EMPLOYMENT AND DECENT WORK FOR ALL.

SDG 8 Decent Work and Economic Growth is moderately addressed by the EU R&I Framework Programme. All parts of the Horizon 2020 Programme, with the exception of the Excellent Science pillar, seem to equally support decent work and economic growth (between 25% to 27% of investment in Pillar 2, 3 and ‘Other’ parts of Horizon 2020 potentially relates to SDG 8). Based on the most frequent key words analysis, several projects relate to labour market and youth unemployment. Many projects interlink with SDG 11 Sustainable Cities and Communities and SDG 12 Climate Action. Based on the latest Global Sustainable Development Repot 201918 the main trade-offs have been so far identified with the SDG 12 Responsible Consumption and Production, SDG 13 Climate Actions and SDG 14 Life Below Water.

TOP 10 MOST ACTIVE ORGANISATIONS % PROJECTS & INVESTMENT % PROJECTS INTERLINKING WITH other SDGs

FRAUNHOFER GESELLSCHAFT ZUR FOERDERUNG DER ANGEWANDTEN 1 FORSCHUNG E.V. CENTRE NATIONAL DE LA RECHERCHE 2 SCIENTIFIQUE CNRS COMMISSARIAT A L ENERGIE ATOMIQUE 3 ET AUX ENERGIES ALTERNATIVES

4 CONSIGLIO NAZIONALE DELLE RICERCHE FUNDACION TECNALIA RESEARCH & 5 INNOVATION ETHNIKO KENTRO EREVNAS KAI 6 TECHNOLOGIKIS ANAPTYXIS

7 TEKNOLOGIAN TUTKIMUSKESKUS VTT OY NEDERLANDSE ORGANISATIE VOOR TOEGEPAST NATUURWETENSCHAPPELIJK 8 ONDERZOEK TNO

9 KATHOLIEKE UNIVERSITEIT LEUVEN

10 UNIVERSITY COLLEGE LONDON

WORD CLOUD - MOST SIGNIFICANT KEY WORDS USED IN HORIZON 2020 PROJECTS RELATED TO SDG 8

ADRESSING THE SDG 8 IN HORIZON 2020: WIN-WIN STRATEGIES FOR GREEN GROWTH The GREEN-WIN project applied a solution-oriented research approach to understand the links between climate action, sustainability, and overcoming implementation barriers through win-win strategies. The project critically assessed where and when win-win and green growth strategies work in practice and where fundamental trade-offs must be faced. It also produced a booklet of green-win narratives concerning sustainability finance and policies contributions for overcoming economic challenges to de-carbonization.

18 See Box 1-2 Assessing interaction among Sustainable Development Goals

UP TO 28% OF HORIZON 2020 INVESTMENT (EUR 12.5 BILLION) POTENTIALLY RELATES TO BUILDING RESILIENT INFRASTRUCTURE, PROMOTE SUSTAINABLE INDUSTRIALISATION AND FOSTER INNOVATION

SDG 9 Industry, Innovation and Infrastructure is moderately addressed by the EU R&I Framework Programme. The highest share of industry, innovation and infrastructure related projects is identified in the Industrial Leadership and Societal Challenges part of the Horizon 2020 Programme (around 40% of Horizon 2020 investment in this pillars potentially relates to SDG 9). Based on the most frequent key words analysis many projects seem related to transport. The majority of projects also interlink with Climate Action (80%), Sustainable Cities and Communities (72%) and Affordable and Clean Energy (70%). Based on the latest Global Sustainable Development Repot 201919 the main trade-offs have been so far identified with Reduced Inequalities (SDG 10), Life on Land (SDG 15) and Life Below Water (SDG 14).

TOP 10 MOST ACTIVE ORGANISATIONS % PROJECTS & INVESTMENT % PROJECTS INTERLINKING WITH other SDGs

FRAUNHOFER GESELLSCHAFT ZUR 1 FOERDERUNG DER ANGEWANDTEN FORSCHUNG E.V. COMMISSARIAT A L ENERGIE ATOMIQUE 2 ET AUX ENERGIES ALTERNATIVES CENTRE NATIONAL DE LA RECHERCHE 3 SCIENTIFIQUE CNRS 4 CONSIGLIO NAZIONALE DELLE RICERCHE 5 TEKNOLOGIAN TUTKIMUSKESKUS VTT OY FUNDACION TECNALIA RESEARCH & 6 INNOVATION ETHNIKO KENTRO EREVNAS KAI 7 TECHNOLOGIKIS ANAPTYXIS NEDERLANDSE ORGANISATIE VOOR 8 TOEGEPAST NATUURWETENSCHAPPELIJK ONDERZOEK TNO 9 POLITECNICO DI MILANO

10 ATOS SPAIN SA

WORD CLOUD - MOST SIGNIFICANT KEY WORDS USED IN HORIZON 2020 PROJECTS RELATED TO SDG 9

ADRESSING THE SDG 9 IN HORIZON 2020: BIOMETRIC ONE-TIME PASSWORD CARD ENABLING SECURE E-TRANSACTIONS QuardCard developed a smart card solution for encrypted communication that enables secure e-banking, e-commerce and e-government using biometric one-time passwords.

19 See Box 1-2 Assessing interaction among Sustainable Development Goals

UP TO 22% OF HORIZON 2020 INVESTMENT (EUR 9.9 BILLION) POTENTIALLY RELATES TO REDUCING INEQUALITIES WITHIN AND AMONG COUNTRIES.

SDG 10 Reduced Inequalities is moderately addressed by the EU R&I Framework Programme. The highest share of projects potentially related to reducing inequalities is identified in the ‘top-down’ parts of the Horizon 2020 Programme related to Societal Challenges (up to 36% of Horizon 2020 investment in this pillar potentially relates to SDG 10). Based on the most frequent key words analysis, this portfolio of projects seems to have a particular focus on migrants and refugees, political and institutional governance and social innovation. Reduced inequality labelled projects commonly link with the Goals of the Climate Action, Good Health and Well-Being and Peace, Justice and Strong Institutions. Based on the latest Global Sustainable Development Repot 201920 the main trade-offs were so far identified with the Affordable and Clean Energy (SDG 7), Decent Work and Economic Growth (SDG 8) and Industry, Innovation and Infrastructure (SDG 9).

TOP 10 MOST ACTIVE ORGANISATIONS % PROJECTS & INVESTMENT % PROJECTS INTERLINKING WITH other SDGs FRAUNHOFER GESELLSCHAFT ZUR 1 FOERDERUNG DER ANGEWANDTEN FORSCHUNG E.V. CENTRE NATIONAL DE LA RECHERCHE 2 SCIENTIFIQUE CNRS 3 CONSIGLIO NAZIONALE DELLE RICERCHE 4 UNIVERSITY COLLEGE LONDON 5 KATHOLIEKE UNIVERSITEIT LEUVEN ETHNIKO KENTRO EREVNAS KAI 6 TECHNOLOGIKIS ANAPTYXIS 7 KOBENHAVNS UNIVERSITET AGENCIA ESTATAL CONSEJO SUPERIOR 8 DEINVESTIGACIONES CIENTIFICAS THE CHANCELLOR, MASTERS AND 9 SCHOLARS OF THE UNIVERSITY OF OXFORD

10 UNIVERSITEIT UTRECHT

WORD CLOUD - MOST SIGNIFICANT KEY WORDS USED IN HORIZON 2020 PROJECTS RELATED TO SDG 10

ADRESSING THE SDG 10 IN HORIZON 2020: INTERNATIONAL NETWORK FOR COMPARING SOCIAL INEQUALITIES The aim of the INCASI project is to create an International Network for Comparative Analysis of Social Inequalities (INCASI) with 19 universities, 10 from Europe and 9 from Latin America and impact academic and policy debates on the subject.

20 See Box 1-2 Assessing interaction among Sustainable Development Goals

UP TO 43% OF HORIZON 2020 INVESTMENT (EUR 19.5 BILLION) POTENTIALLY RELATES TO MAKING CITIES AND HUMAN SETTLEMENTS INCLUSIVE, SAFE, RESILIENT AND SUSTAINABLE.

SDG 11 Sustainable Cities and Communities is well addressed by the EU R&I Framework Programme. A relatively high share (from 48% to 58%) of investment is potentially related to Sustainable Cities and Communities in all parts of Horizon 2020, with exception of Pillar 1 Excellence Science where the share is much lower (19%). Based on the most frequent key words analysis, these projects often refer to nature based solutions and follower cities. The majority of projects interlink with Climate Action Goal (78%). Based on the latest Global Sustainable Development Repot 201921 the main trade-offs have been so far identified with SDG 14 Life Below Water and SDG 15 Life on Land.

TOP 10 MOST ACTIVE ORGANISATIONS % PROJECTS & INVESTMENT % PROJECTS INTERLINKING WITH other SDGs

FRAUNHOFER GESELLSCHAFT ZUR 1 FOERDERUNG DER ANGEWANDTEN FORSCHUNG E.V. CENTRE NATIONAL DE LA RECHERCHE 2 SCIENTIFIQUE CNRS 3 CONSIGLIO NAZIONALE DELLE RICERCHE COMMISSARIAT A L ENERGIE ATOMIQUE 4 ET AUX ENERGIES ALTERNATIVES 5 TEKNOLOGIAN TUTKIMUSKESKUS VTT OY ETHNIKO KENTRO EREVNAS KAI 6 TECHNOLOGIKIS ANAPTYXIS AGENCIA ESTATAL CONSEJO SUPERIOR 7 DEINVESTIGACIONES CIENTIFICAS NEDERLANDSE ORGANISATIE VOOR 8 TOEGEPAST NATUURWETENSCHAPPELIJK ONDERZOEK TNO 9 KATHOLIEKE UNIVERSITEIT LEUVEN FUNDACION TECNALIA RESEARCH & 10 INNOVATION

WORD CLOUD - MOST SIGNIFICANT KEY WORDS USED IN HORIZON 2020 PROJECTS RELATED TO SDG 11

ADRESSING THE SDG 11 IN HORIZON 2020: URBAN NATURE LABS UNaLab, Urban Nature Labs, is developing a European framework of replicable, locally attuned nature-based solutions to enhance the climate and water resilience of cities together with stakeholders and citizens. In 10 cities around the world, the UNaLab project implements local solutions such as urban ecological water management, accompanied with greening measures and innovative, inclusive urban design.

21 See Box 1-2 Assessing interaction among Sustainable Development Goals

100

UP TO 28% OF HORIZON 2020 INVESTMENT (EUR 12.5 BILLION) POTENTIALLY RELATES TO ENSURING SUSTAINABLE CONSUMPTION AND PRODUCTION PATTERNS

SDG 12 Responsible Consumption and Production is moderately addressed by the EU R&I Framework Programme. From 31% to 42% of investment is potentially related to Responsible Consumption and Production in all parts of Horizon 2020, with exception of Pillar 1 Excellence Science where the share is much lower (10%). Based on the most frequent key words analysis, project reports often mention circular economy, food waste and waste management in general. The majority of projects (88%) interlink with the Climate Action Goal. Based on the latest Global Sustainable Development Repot 201922 acting on Responsible Consumption and Production brings co-benefits across the other SDGs, with the exception of SDG 8 on Decent Work on Economic Growth, where some trade-offs are evidenced.

TOP 10 MOST ACTIVE ORGANISATIONS % PROJECTS & INVESTMENT % PROJECTS INTERLINKING WITH other SDGs

FRAUNHOFER GESELLSCHAFT ZUR 1 FOERDERUNG DER ANGEWANDTEN FORSCHUNG E.V. CENTRE NATIONAL DE LA RECHERCHE 2 SCIENTIFIQUE CNRS 3 CONSIGLIO NAZIONALE DELLE RICERCHE COMMISSARIAT A L ENERGIE ATOMIQUE 4 ET AUX ENERGIES ALTERNATIVES AGENCIA ESTATAL CONSEJO SUPERIOR 5 DEINVESTIGACIONES CIENTIFICAS 6 TEKNOLOGIAN TUTKIMUSKESKUS VTT OY FUNDACION TECNALIA RESEARCH & 7 INNOVATION INSTITUT NATIONAL DE LA RECHERCHE 8 AGRONOMIQUE 9 STICHTING WAGENINGEN RESEARCH 10 POLITECNICO DI MILANO

WORD CLOUD - MOST SIGNIFICANT KEY WORDS USED IN HORIZON 2020 PROJECTS RELATED TO SDG 12

ADRESSING THE SDG 12 IN HORIZON 2020: DISCARDED TEXTILES AS RAW MATERIAL RESYNTEX project established a pilot textile recycling plant in Slovenia to create a new circular economy model for the textile and chemical industries. Recovering secondary raw materials from unwearable textile waste, the installation covers the whole value chain.

22 See Box 1-2 Assessing interaction among Sustainable Development Goals

101

UP TO 54% OF HORIZON 2020 INVESTMENT (EUR 24 BILLION) POTENTIALLY RELATES TO REDUCING FUTURE GREENHOUSE GAS EMISSIONS.

SDG 13 Climate Action is the most well addressed Goal by the EU R&I Framework Programme. The highest share of climate action related projects is identified in the ‘top-down’ parts of the Horizon 2020 Programme related to Societal Challenges (up to 67% of Horizon 2020 investment in this pillar potentially relates to SDG 13). Based on the most frequent key words analysis, projects relate to climate science including climate projections, earth and system observations and assessment. Almost half of the Climate Action projects interlink with affordable and clean energy (SDG 07), sustainable cities and communities (SDG 11) and good health and well-being (SDG 03). Based on the latest Global Sustainable Development Repot 201923 these four SDGs have a potential to harness co-benefits (i.e. by tackling one SDG positive effects are expected on the other). The trade-offs, however lie especially with the SDG 08 on decent work and economic growth and SDG 02 on zero poverty. A much lover share of current Horizon 2020 investment seems to be addressing these trade-offs.

TOP 10 MOST ACTIVE ORGANISATIONS % PROJECTS & INVESTMENT % PROJECTS INTERLINKING WITH other SDGs CENTRE NATIONAL DE LA RECHERCHE 1 SCIENTIFIQUE CNRS FRAUNHOFER GESELLSCHAFT ZUR 2 FOERDERUNG DER ANGEWANDTEN FORSCHUNG E.V. 3 CONSIGLIO NAZIONALE DELLE RICERCHE COMMISSARIAT A L ENERGIE ATOMIQUE 4 ET AUX ENERGIES ALTERNATIVES AGENCIA ESTATAL CONSEJO SUPERIOR 5 DEINVESTIGACIONES CIENTIFICAS THE CHANCELLOR MASTERS AND 6 SCHOLARS OF THE UNIVERSITY OF CAMBRIDGE 7 KATHOLIEKE UNIVERSITEIT LEUVEN 8 UNIVERSITY COLLEGE LONDON 9 DANMARKS TEKNISKE UNIVERSITET

10 TEKNOLOGIAN TUTKIMUSKESKUS VTT OY

WORD CLOUD - MOST SIGNIFICANT KEY WORDS USED IN HORIZON 2020 PROJECTS RELATED TO SDG 13

ADRESSING THE SDG 13 IN HORIZON 2020: REDUCING THE CARBON FOOTPRINT OF SHIPPING The German project AtlantOS is computing optimal shipping tracks in the Atlantic Ocean to achieve reduction of carbon intensity through ship route optimization.

23 See Box 1-2 Assessing interaction among Sustainable Development Goals

102

UP TO 32% OF HORIZON 2020 INVESTMENT (EUR 14.2 BILLION) POTENTIALLY RELATES TO CONSERVING AND SUSTAINABLY USING THE OCEANS, SEAS AND MARINE RESOURCES FOR SUSTAINABLE DEVELOPMENT.

SDG 14 Life Below Water is moderately addressed by the EU R&I Framework Programme. The highest share of Life Below Water related projects is identified in the ‘top-down’ parts of the Horizon 2020 Programme related to Societal Challenges (up to 42% of Horizon 2020 investment in this pillar potentially relates to SDG 14). Based on the most frequent key words analysis, these projects frequently reference marine ecosystems, coral reefs, fisheries and the overall Blue growth strategy. The majority of projects (78%) interlink with the Climate Action or Life on Land (66%). Based on the latest Global Sustainable Development Repot 201924 the trade-offs lie especially with 01 No Poverty, 02 Zero Hunger, 13 Climate Action and 09 Industry, Innovation and Infrastructure.

TOP 10 MOST ACTIVE ORGANISATIONS % PROJECTS & INVESTMENT % PROJECTS INTERLINKING WITH other SDGs

CENTRE NATIONAL DE LA RECHERCHE 1 SCIENTIFIQUE CNRS FRAUNHOFER GESELLSCHAFT ZUR 2 FOERDERUNG DER ANGEWANDTEN FORSCHUNG E.V. AGENCIA ESTATAL CONSEJO SUPERIOR 3 DEINVESTIGACIONES CIENTIFICAS 4 CONSIGLIO NAZIONALE DELLE RICERCHE COMMISSARIAT A L ENERGIE ATOMIQUE 5 ET AUX ENERGIES ALTERNATIVES UNITED KINGDOM RESEARCH AND 6 INNOVATION INSTITUT NATIONAL DE LA RECHERCHE 7 AGRONOMIQUE 8 KOBENHAVNS UNIVERSITET MAX-PLANCK-GESELLSCHAFT ZUR 9 FORDERUNG DER WISSENSCHAFTEN EV 10 STICHTING WAGENINGEN RESEARCH

WORD CLOUD - MOST SIGNIFICANT KEY WORDS USED IN HORIZON 2020 PROJECTS RELATED TO SDG 14

ADRESSING THE SDG 14 IN HORIZON 2020: INCREASING OCEAN LITERACY Respon-SEA-ble co-designed, created, tested and assessed a range of Ocean Literacy tools and products. The project made specific suggestions to maximise opportunities in existing policies (such as the Marine Strategy Framework Directive) to strengthen Ocean Literacy of different groups.

24 See Box 1-2 Assessing interaction among Sustainable Development Goals

103

UP TO 32% OF HORIZON 2020 INVESTMENT (EUR 15.6 BILLION) POTENTIALLY RELATES TO PROTECTING, RESTORING AND PROMOTING SUSTAINABLE USE OF TERRESTRIAL ECOSYSTEMS, SUSTAINABLY MANAGING FORESTS, COMBAT DESERTIFICATION, AND HALTING AND REVERSING LAND DEGRADATION AND HALTING BIODIVERSITY LOSS.

SDG 15 Life on Land is moderately addressed by the EU R&I Framework Programme. An important share of Life on Land related investment is estimated across the different parts of the Horizon 2020 Programme (from 23% of investment in Pillar 1 Excellent Science to up to 42% in Pillar 3 Societal Challenges). Based on the most frequent key words analysis, these projects frequently mention ecosystem services, forest and land management as well as biodiversity. The majority of projects (75%) interlink with SDG 13 on Climate Action. Based on the latest Global Sustainable Development Repot 201925 trade-offs lie in particular with the SDG 7 Affordable and Clean Energy and SDG 1 No Poverty and SDG 2 Zero Hunger.

TOP 10 MOST ACTIVE ORGANISATIONS % PROJECTS & INVESTMENT % PROJECTS INTERLINKING WITH other SDGs CENTRE NATIONAL DE LA RECHERCHE 1 SCIENTIFIQUE CNRS FRAUNHOFER GESELLSCHAFT ZUR 2 FOERDERUNG DER ANGEWANDTEN FORSCHUNG E.V. 3 CONSIGLIO NAZIONALE DELLE RICERCHE AGENCIA ESTATAL CONSEJO SUPERIOR 4 DEINVESTIGACIONES CIENTIFICAS INSTITUT NATIONAL DE LA RECHERCHE 5 AGRONOMIQUE COMMISSARIAT A L ENERGIE ATOMIQUE 6 ET AUX ENERGIES ALTERNATIVES 7 KOBENHAVNS UNIVERSITET UNITED KINGDOM RESEARCH AND 8 INNOVATION 9 STICHTING WAGENINGEN RESEARCH

10 WAGENINGEN UNIVERSITY

WORD CLOUD - MOST SIGNIFICANT KEY WORDS USED IN HORIZON 2020 PROJECTS RELATED TO SDG 15

ADRESSING THE SDG 15 IN HORIZON 2020: ALGAE-BASED FERTILISER VegaAlga developed a microalgae-based fertiliser to establish a sustainable agricultural ecosystem to meet the growing demand for ‘green’ vegetables. The team developed a smaller microalgae-fertiliser production system to allow farmers to produce fertiliser on their own land in a cost-effective and eco-friendly way.

25 See Box 1-2 Assessing interaction among Sustainable Development Goals

104

UP TO 31% OF HORIZON 2020 INVESTMENT (EUR 14 BILLION) POTENTIALLY RELATES TO PROMOTING PEACEFUL AND INCLUSIVE SOCIETIES FOR SUSTAINABLE DEVELOPMENT, PROVIDING ACCESS TO JUSTICE FOR ALL AND BUILDING EFFECTIVE, ACCOUNTABLE AND INCLUSIVE INSTITUTIONS AT ALL LEVELS.

SDG 16 Peace, Justice and Strong Institutions is moderately addressed by the EU R&I Framework Programme. The highest share of Peace, Justice and Strong Institutions related projects is identified in the ‘top-down’ parts of the Horizon 2020 Programme related to Societal Challenges (up to 42% of Horizon 2020 investment in this pillar potentially relates to SDG 16). Based on the most frequent key words analysis, these projects mention Law Enforcement Agencies (LEAs), radicalisation and organised crime. More than half of the projects interlink with the Good Health and Well-Being (SDG 03), Climate Action (SDG 13) and Sustainable Cities and Communities (SDG 11). Based on the latest Global Sustainable Development Repot 201926 there is little knowledge about the co-benefits and trade-offs - trade-offs were so far identified with Good Health and Well-Being (SDG 03) and Affordable and Clean Energy (SDG 07).

TOP 10 MOST ACTIVE ORGANISATIONS % PROJECTS & INVESTMENT % PROJECTS INTERLINKING WITH other SDGs FRAUNHOFER GESELLSCHAFT ZUR 1 FOERDERUNG DER ANGEWANDTEN FORSCHUNG E.V. CENTRE NATIONAL DE LA RECHERCHE 2 SCIENTIFIQUE CNRS 3 CONSIGLIO NAZIONALE DELLE RICERCHE 4 UNIVERSITY COLLEGE LONDON 5 KATHOLIEKE UNIVERSITEIT LEUVEN ETHNIKO KENTRO EREVNAS KAI 6 TECHNOLOGIKIS ANAPTYXIS COMMISSARIAT A L ENERGIE ATOMIQUE 7 ET AUX ENERGIES ALTERNATIVES THE CHANCELLOR, MASTERS AND 8 SCHOLARS OF THE UNIVERSITY OF OXFORD 9 ATOS SPAIN SA AGENCIA ESTATAL CONSEJO SUPERIOR 10 DEINVESTIGACIONES CIENTIFICAS

WORD CLOUD - MOST SIGNIFICANT KEY WORDS USED IN HORIZON 2020 PROJECTS RELATED TO SDG 16

ADRESSING THE SDG 16 IN HORIZON 2020: POLICIES FOR PEACE POLICIES_FOR_PEACE project aims to study which key institutions and policies are best suited to reduce incentives for engaging in appropriation and armed conflict.

26 See Box 1-2 Assessing interaction among Sustainable Development Goals

105

UP TO 2% OF HORIZON 2020 INVESTMENT (EUR 1 BILLION) POTENTIALLY RELATES TO REVITALISING THE GLOBAL PARTNERHIP FOR SUSTAINABLE DEVELOPMENT

SDG 17 Partnerships for the Goals is the least addressed Goal by the EU R&I Framework Programme. A small share of Horizon 2020 investment (2%) potentially relates to revitalising the global partnership for sustainable development which focuses on enhancing the North-South and South-South cooperation, promoting international trade and helping developing countries to increase their exports. Based on the latest Global Sustainable Development Repot 201927 trade-offs are limited and were so far identified only with SDG 4 Quality Education and SDG 16 Peace, Justice and Strong Institutions.

TOP 10 MOST ACTIVE ORGANISATIONS % PROJECTS & INVESTMENT % PROJECTS INTERLINKING WITH other SDGs

FRAUNHOFER GESELLSCHAFT ZUR 1 FOERDERUNG DER ANGEWANDTEN FORSCHUNG E.V. DEUTSCHES ZENTRUM FUER LUFT - UND 2 RAUMFAHRT EV 3 STICHTING WAGENINGEN RESEARCH 4 CONSIGLIO NAZIONALE DELLE RICERCHE CENTRE NATIONAL DE LA RECHERCHE 5 SCIENTIFIQUE CNRS INSTITUT NATIONAL DE LA RECHERCHE 6 AGRONOMIQUE 7 NORGES FORSKNINGSRAD UNITED KINGDOM RESEARCH AND 8 INNOVATION 9 UNIVERSITY COLLEGE LONDON FUNDACAO PARA A CIENCIA E A 10 TECNOLOGIA

WORD CLOUD - MOST SIGNIFICANT KEY WORDS USED IN HORIZON 2020 PROJECTS RELATED TO SDG 17

ADRESSING THE SDG 17 IN HORIZON 2020: ACCELERATING OPEN IOT AND BIG DATA INNOVATION IN AFRICA WAZIHUB (in Swahili for Open-Hub) is an innovation project for Africa aiming to create an OpenHUB of IoT and Big data cutting-edge African-grade solutions, co-designed in Africa to match local service needs.

27 See Box 1-2 Assessing interaction among Sustainable Development Goals

106

ANNEX SDG Related Key Words

SDG IRIS Query

"alleviate poverty" OR "breadline" OR "disaster control" OR "disaster management" OR "disaster prevention" OR "economic marginalisation" OR "end poverty" OR "eradicate poverty" OR "extreme poverty" OR "financial inclusion" OR "income equality" OR "income gap" OR "living in poverty" OR "living standard" OR "microfinancing" OR "poor and vulnerable" OR "poverty alleviation" OR "poverty line" OR "poverty" OR 01 POVERTY "precarious situation" OR "purchasing power parity" OR "resilience climate change" OR "resilience disasters" OR "secure livelihood" OR "social exclusion" OR "social protection" OR "social protection system" OR "vulnerability climate" OR "vulnerability disaster" OR "wealth distribution" OR "working poor"

"agricultur* practice" OR "agricultur* production " OR "agricultur* productivity " OR "agricultur* resilient " OR " agricultur* subsidies " OR " agricultur* policy " OR " agricultur* waste " OR " agricultur* adaptation " OR " agricultur* and climate change " OR " agricultur* biodiversity " OR " agricultur* rural " OR " child* hunger " OR " child* hungry " OR " child* stunted growth " OR " child* stunting " OR " child* wasting " OR " conservation facilit* animals " OR " conservation facilit* plants " OR " conservation facilit* seeds " OR " conservation facilit* species " OR " crop diversity " OR " doha development round " OR " ecological seed " OR " food commodity market " OR " food 02 HUNGER insecurity " OR " food needs " OR " food price anomal* " OR " food price volatility " OR " food reserve " OR " food security " OR " food sufficient " OR " food sovereignty " OR " food waste " OR " fair trade " OR " gene* diversity* animals " OR " gene* diversity* plants " OR " gene* diversity* seeds " OR " gene* diversity* species " OR " global food " OR " hunger " OR " malnutrition " OR " nutriti* food " OR " nutriti* improved " OR " nutriti* poor " OR " nutrition* insecurity " OR " nutrition* needs " OR " nutrition* security " OR " plant bank " OR " livestock gene bank " OR " quality land " OR " quality soil " OR " small farm " OR " small-scale food producers " OR " sustain* agricultur* " OR " sustain* food production " OR " secure livelihood " OR " trade restriction " OR " Undernourish* "

" AIDS " OR " adolescent birth rate " OR " adolescent mother " OR " affordable health* " OR " affordable medicin* " OR " alcohol abuse " OR " alcohol addict* " OR " antibiotic resistance " OR " bipolar disorder " OR " chronic depression " OR " contraception " OR " disease " OR " drug abuse " OR " drug abuse " OR " drug addict* " OR " Ebola " OR " elderly care " OR " epidemic " OR " family planning " OR " global health " OR " global health risk " OR " health coverage " OR " health financ* " OR " health risk " OR " health training* " OR " health-care services " OR " hepatitis " OR " HIV " OR " immunization* " OR " infant mortality " OR " insulin resistance " OR " like expectancy " OR " malaria " OR " 03 HEALTH maternal mortality " OR " medicine reproductive " OR " medicines " OR " mental disorder " OR " mental health " OR " monopolar disorder " OR " mortality antenatal " OR " mortality before five " OR " mortality child " OR " mortality neonatal " OR " mortality premature " OR " mortality under five " OR " nursing " OR " patient care " OR " pharmacy " OR " premature death " OR " psychiatric disorder " OR " public health " OR " reproductive health* " OR " road traffic accident* " OR " road traffic death " OR " road traffic injur* " OR " sexual health* " OR " sexual infection " OR " substance abuse " OR " substance abuse " OR " substance addiction " OR " teen* pregnan* " OR " tobacco " OR " tropical disease " OR " tuberculosis " OR " vaccin* " OR " Zika "

" access to education " OR " child care " OR " develop* countr* education " OR " early childhood education " OR " equitable education " OR " free education " OR " global funding of education " OR " inclusive education " OR " inclusive learning environment " OR " increase skills " OR 04 EDUCATION " learning environment " OR " learning opportunities " OR " lifelong learning " OR " literacy " OR " numeracy " OR " pre-primary education " OR " primary education " OR " primary school " OR " qualified teachers " OR (" refugees" AND "learning ") OR " school enrolment " OR " secondary school " OR " teacher training " OR " uneducated " OR " universal education "

" child marriage " OR " equal pay " OR " feminism " OR " forced marriage " OR " gender *equalit* " OR " gender budgeting " OR " gender differences " OR " gender discrimination " OR " genital cutting " OR " genital mutilation " OR " governance and gender " OR " human 05 GENDER trafficking " OR " reproductive rights " OR " sexual and reproductive health " OR " sexual discrimination " OR " sexual violence " OR " unpaid care " OR " unpaid domestic work " OR " violence against girls " OR (" violence against women ") OR " women at work " OR " women* empower* " OR " women* rights "

" wastewater treatment " OR " wetlands " OR " desalination " OR " aquifers " OR " hygiene " OR " sustainable water governance " OR " defecation " OR " sanitation " OR " urban water " OR " drinking water " OR " untreated wastewater " OR " water footprint " OR " water 06 WATER ecosystem " OR " water quality " OR " water scarcity " OR " water harvesting " OR " water availability " OR " water pollution " OR " water supply " OR " water disaster* " OR " water-use efficiency " OR " wastewater " OR " hydropower " OR " irrigation " OR " lakes " OR " sanitation management " OR " sewerage " OR " water resource management " OR " toilet* "

" alternative energy " OR " clean energy " OR " clean fuel " OR " decentralised energy supply " OR " e-City " OR " electric car " OR " e-mobility " OR " energy consumption " OR " energy efficiency " OR " energy independence " OR " energy infrastructure " OR " energy intensity " OR " energy mix " OR " energy safe " OR " energy self-sufficient " OR " energy service* " OR " energy source* " OR " energy storage " OR " energy 07 ENERGY technology " OR " energy transition " OR " future city " OR " hydropower " OR " passive house " OR " photovoltaics " OR " plus-energy " OR " renewable energy " OR " smart grid " OR " solar thermic " OR " sustainability assessment " OR " sustainable energy " OR " waste energy " OR " waste renewable " OR " wind power " OR " zero energy "

" child labour " OR " decent job " OR " decent work " OR " economic productivity " OR " entrepreneurship " OR " equal pay " OR " forced labour " OR " gainful employment " OR " human trafficking " OR " inclusive economic growth " OR " industrialisation " OR " job creation " OR " labour market " OR " labour right " OR " low wage " OR " microfinancing " OR " modern slavery " OR " secure working environment " OR " 08 WORK & GROWTH sustainable consumption " OR " sustainable economic growth " OR " sustainable production " OR " sustainable tourism " OR " trade support " OR " work opportunities " OR " work safety " OR " youth employment " OR " youth unemployment " OR " child soldiers " OR " migrant workers " OR " stable employment " OR " financial service " OR " informal employment "

" alternative fuel " OR " rail network " OR " affordable credit " OR " resource use efficiency " OR " transborder infrastructure " OR " combined traffic " OR " efficient communication network " OR " sustainable infrastructure " OR " sustainable industrialization " OR " sustainable information technologies " OR " sustainable communication technologies " OR " sustainable transport " OR " sustainability assessment " OR 09 INDUSTRY & " regional infrastructure " OR " resilient infrastructure " OR " resilient towards disasters " OR " transformation sustainability " OR " traffic INNOVATION modal shift " OR " waterway " OR " clean technologies " OR " environmentally sound technologies " OR " financial service " OR " ICT infrastructure " OR " Industrial diversification " OR " industrialisation " OR " Information and communication technology " OR " Network infrastructure " OR " Resource use efficiency " OR " industry value chains " OR " water infrastructure " OR " inclusive industry " OR " manufacturing industry " OR " high-tech industry "

107

" alleviate poverty " OR " developing countries decision-making " OR " empowerment " OR " equal opportunities " OR " fiscal protection " OR " human rights " OR " income equality " OR " income gap " OR " income growth " OR " indigenous " OR " international law " OR " least 10 INEQUALITY developed countries " OR " microfinancing " OR " migration policy " OR " open society " OR " parallel society " OR " poverty " OR " racism " OR " remittance " OR " resilient infrastructure " OR " small island developing states " OR " social contract " OR " basic income " OR " vulnerable nations " OR " wage protection " OR " development aid "

" air quality "OR " bio waste "OR " city development "OR " city planning "OR " city pollution "OR " community building "OR " e-City "OR " energy recovery "OR " facade system "OR " green space "OR " heat island "OR " intercultural garden "OR " landscape water engineering "OR

" mobilit " OR " passive house "OR " resilience in regions "OR " resilient building "OR " resource recovery "OR " risk reduction "OR " safe city © European Union, 2020 "OR " safe housing "OR " sharing economy "OR " smart city "OR " suburban "OR " sustainability assessment "OR " sustainability model "OR " sustainable building design "OR " sustainable cities "OR " housing "OR " sustainable living "OR " sustainable properties "OR " sustainable 11 CITIES residential location choice "OR " sustainable settlement "OR " sustainable transport "OR " sustainable urbanization "OR " town planning "OR " transition town "OR " urban development "OR " urban gardening "OR " urban lab "OR " urban mining "OR " urban planning "OR " urban resilience "OR " urban transport "OR " waste collection "OR " waste disposal "OR " waste industry "OR " waste management "OR " water management "OR " waste logistics "OR " world heritage "OR " Disaster risk reduction "OR " over crowding "OR " public space "OR " public transport "OR " road safety "OR " slums "OR " smart cities "OR " urbanisations "

" waste prevention "OR " waste reduction "OR " resource efficiency "OR " life cycle analysis "OR " local products "OR " sustainable production "OR " sustainable resources "OR " sustainable supply "OR " sustainable consumption "OR " sustainable management "OR " sustainable tourism "OR " sustainable forest management "OR " sustainable economy "OR " sustainability report "OR " sustainability assessment "OR " 12 RESPONSIBLE sustainability certificate "OR " food loss "OR " food waste "OR " Obsolescence "OR " resource management "OR " optimal resource planning CONSUMPTION "OR " environmental management systems "OR " recycling "OR " corporate sustainable responsibility "OR " overconsumption "OR " sustainable agriculture "OR " sustainable service design "OR " sustainable product design "OR " sustainable infrastructure "OR " postharvest loss "OR " sustainable procurement "OR " sustainable buying "OR " chemical waste "OR " responsible production chains "OR " material footprint "OR " natural resources "

" air quality " OR " albedo " OR " carbon dioxide " OR " CFC " OR " climate change " OR " climate protection " OR " early signals " OR " early warning " OR " environment impact " OR " environment pollution " OR " environment protection " OR " environment stress " OR " global temperature " OR " global warming " OR " greenhouse effect " OR " greenhouse gas " OR " heat island " OR " Ice loss " OR " low-carbon 13 CLIMATE economy " OR " methane " OR " natural disasters " OR " natural hazard " OR " natural systems " OR " ocean warming " OR " ocean warming " OR " ozone hole " OR " Paris Agreement " OR " peak oil " OR " Pollution " OR " Renewable " OR " sea level rise " OR " rising sea " OR " sustainability assessment " OR " sustainable development " OR " climate policy "

" aquaculture " OR " Biodiversity " OR " fishing " OR " fisheries subsidies " OR " small-scale artisanal fisheries " OR " coral reef " OR " coral bleaching " OR " marine area " OR " marine resources " OR " marine technology " OR " marine pollution " OR " nutrient pollution " OR " overfishing " OR " ocean acidification " OR " coastal ecosystem " OR " costal resources " OR " fish species " OR " sustainable ecosystems " 14 LIFE IN WATER OR " ocean " OR " seawater " OR " ocean protection " OR " marine protection " OR " marine preservation " OR " marine conservation " OR " environmental degradation " OR " eutrophication " OR " acidification " OR " fisheries " OR " fish stock " OR " marine law " OR " fish* policy " OR " Ocean temperature " OR " Oceanography " OR " Sea grasses "

" afforestation " OR " biodiversity " OR " deforestation " OR " desertification " OR " drought " OR " ecosystem " OR " ecosystem services " OR " humification " OR " invasive species " OR " land degradation " OR " land ecosystem " OR " mountain ecosystem " OR " pedogenesis " OR " poaching " OR " protected areas " OR " Protected fauna " OR " Protected flora " OR " protected species " OR " Reforestation " OR " soil 15 LIFE ON LAND degradation " OR " forest management " OR " mountain development " OR " sustainable land use " OR " sustainable soil management " OR " terrestrial ecosystem " OR " threatened species " OR " wetlands " OR " natural habitat " OR " wildlife " OR " conservation " OR " nagoya protocol "

" arm trafficking " OR " arms flow " OR " arms trade " OR " armed conflict " OR " civil conflict " OR " civil war " OR " bribery " OR " child abuse " OR " child exploitation " OR " child torture " OR " child trafficking " OR " child violence " OR " constitutionality " OR " corruption " OR " crime " OR " e-democracy " OR " fundament freedom " OR " human rights " OR " human trafficking " OR " international law " OR " justice " OR " legal 16 PEACE, JUSTICE AND identity " OR " national security " OR " non-discriminatory " OR " Non-violence " OR " organized crime " OR " Peace " OR " Peaceful societies " STRONG INSTITUTIONS OR " Physical abuse " OR " Police " OR " Psychological abuse " OR " Rule of law " OR " Security threats " OR " Sexual abuse " OR " Sexual violence " OR " Stolen assets " OR " Tax evasion " OR " terrorism " OR " Theft " OR " Torture " OR " Trafficking " OR " Transparency " OR " Un- sentenced detainees " OR " Unstable societies " OR " Victims of violence " OR " Violence " OR " Violence against women and children " OR " Weapon seizures " OR " conflict management " OR " homicide " OR " ethnic conflict " OR " organised crime "

(" development assistance " OR " developing country " OR " developing countries " ) AND (" European Development Fund " OR " global partnership " OR " investment promotion regimes " OR " capacity-building developing countries " OR " official development assistance " OR 17 PARTNERSHIP " strategic partnership " OR " debt restructuring developing countries " OR " Science cooperation agreement " OR " exports developing countries " OR " World Trade Organization " OR " development agenda " OR " debt relief developing countries”

Other monitoring flash reports available here: https://ec.europa.eu/info/publications/horizon-2020-monitoring-flash_en

#1 Country Participation #2 Dynamic Network Analysis #3 International Cooperation #4 Patents in FP

108

Getting in touch with the EU

IN PERSON All over the European Union there are hundreds of Europe Direct information centres. You can find the address of the centre nearest you at:https://europa.eu/european-union/contact_en

ON THE PHONE OR BY EMAIL Europe Direct is a service that answers your questions about the European Union. You can contact this service: – by freephone: 00 800 6 7 8 9 10 11 (certain operators may charge for these calls), – at the following standard number: +32 22999696, or – by email via: https://europa.eu/european-union/contact_en

Finding information about the EU

ONLINE Information about the European Union in all the official languages of the EU is available on the Europa website at: https:// europa.eu/european-union/index_en

EU PUBLICATIONS You can download or order free and priced EU publications from: https://op.europa.eu/en/publications. Multiple copies of free publications may be obtained by contacting Europe Direct or your local information centre (see https://europa.eu/european-union/contact_en)

EU LAW AND RELATED DOCUMENTS For access to legal information from the EU, including all EU law since 1952 in all the official language versions, go to EUR- Lex at: http://eur-lex.europa.eu

OPEN DATA FROM THE EU The EU Open Data Portal (http://data.europa.eu/euodp/en) provides access to datasets from the EU. Data can be downloaded and reused for free, for both commercial and non-commercial purposes. The monitoring flash series ‘Keeping our eyes on the Horizon’ presents detailed analysis of key aspects of the EU R&I framework programme developments in recent years, such as participation by country and its contribution to sustainability issues. It shows that Horizon 2020 is already helping to build a more sustainable future: four in every five projects tie in with specific sustainable development goals. Programme activities are drawing more and more universities and companies into a massive network of over 1.5 million collaborations worldwide. The programme is delivering quality results in the form of patented inventions with above average market value and scientific publications that are quoted more than the world average.

The series is part of an effort to modernise the monitoring and evaluation of the R&I framework programmes. It complements real time data on programme implementation and results, as provided by the Horizon dashboard and the Horizon results platform, and lays the ground for the Horizon 2020 ex post evaluation in 2023 and the Horizon Europe interim evaluation in 2024.

Research and Innovation policy