CORDIS - EU research results

Learning from Innovation in Public Sector Environements

Final Report Summary - LIPSE (Learning from Innovation in Public Sector Environements)

Executive Summary:
The public sector is wrestling with vital challenges which influence its legitimacy, effectiveness and efficiency. Currently, the most visible is the financial crisis in Europe which forces governments to reconsider the ways in which they provide public services. The ability of the public sector to respond to these challenges influences the viability of the public sector and the services that are provided, especially if one acknowledges that the public sector plays an important role in European societies and economies. In addressing these challenges, it is important to look at the innovation capacity of the public sector as well as its capacity to adopt new innovations. The exploration and exploitation of this innovation capacity requires public sector organizations that are able to incorporate and link knowledge, views, interests and resources in their innovation processes in order to achieve outcomes that are considered as appropriate for and by the stakeholders involved. This is the essence of social innovation. As such, social innovation in the public sector refers to the development of innovations, and their diffusion and adoption that are responsive to the needs of citizens, companies, non-governmental organizations and other stakeholders. The articulation of these needs can be viewed as relevant inputs for social innovation processes. Simultaneously, it is important to understand how these needs are organized within the innovation process itself. That is why the social innovation concept emphasizes that innovations are the result of a process of co-creation among relevant actors. Citizens are important actors in these processes, albeit that their position – in contrast to well organized and institutionalized interests and interest groups – is rather weak. Further, the involvement of many stakeholders and the ensuing dialogue makes social innovation processes rather complex and dynamic. This influences the successful governance of these processes in terms of risk selection and management.

The LIPSE project (Learning from Innovation in Public Sector Environments) identified drivers and barriers to successful social innovation in the public sector. Through studying social innovation and co-creation processes in 11 European countries and 7 policy sectors, LIPSE created and disseminated essential knowledge about public innovation. The LIPSE consortium consisted of leading institutions in 11 European countries: Belgium, Denmark, Estonia, France, Germany, Italy, the Netherlands, Spain, Slovakia, Rumania and the United Kingdom. The research was focused on seven policy sectors that are relevant to the EU: social security/welfare services, (mental) health services, sustainable public services, public utilities, (electronic) procurement, information society/ICT (teleworking), urban/rural planning for regeneration. Moreover, attention was given to the role of regional and local authorities.

Project Context and Objectives:
The LIPSE project sought to comparatively identify relevant drivers and barriers to social innovation in 11 European countries, drawing on a team of European public administration scholars from 11universities in 10 countries.
In seven related international research projects, LIPSE mapped and analysed social innovative mechanisms in the public sector to improve social and policy coordination, especially when the public sector is facing the fiscal crisis. The research contributes to our understanding of the impact of social innovation by integrating sectoral and national analyses and to the development of future public sector reform strategies by drawing lessons from past experience, exploring trends and drafting possible scenarios for social innovation. LIPSE consisted of seven related work packages (WP), each implemented by an international team of researchers. The rationale behind the project was to systematically identify relevant drivers and barriers, and to systematically study the outcomes of social innovations, within the European public sector. It is important to assess innovation processes and social innovation outcomes not only by looking at the efficiency and efficacy gains that these social innovations deliver, but also to look at the extent to which other relevant political and societal values (such as equity, fairness, access) are realized. The implication is that, although we can learn from private sector innovation, we had to take into account the specific institutional features of the public sector. These institutional features constitute the context in which social innovation takes place within the public sector.
Overall objectives of the project were:

1. Identify relevant drivers and barriers that explain the success or failure of social innovations in the public sector, and then to give policy recommendations based on the outcomes of the specific research activities in this proposal.
2. Learn from cross-national and cross-sectoral comparisons to understand how social innovation practice convergence or diverge between EU member states.
3. Advise policymakers and researchers on potential future pathways for social innovation in the public sector that can enhance productivity, growth and competitiveness in Europe.

We also identified specific goals to accomplish these overall goals:
a. Analyse how the institutional environments contribute to social innovation and to develop policy recommendations based on the findings.
b. Explain the conditions under which the participation of citizens, who can be considered as weak interests, contribute to successful forms of co-creation, and to develop policy recommendations based on the findings.
c. Analyse if and how recommendations made by audit offices and ombudsmen, that act as intermediaries in voicing the rather weak interests of citizens, non-governmental organizations and companies have been used as inputs in public sector innovation, and to make policy recommendations on how to improve this.
d. Study how the definition and governance of risks in service innovation projects influences social innovation, as well as to generate policy recommendations.
e. Determine what factors contribute to the successful upscaling of social innovations as well as to give policy guidelines that stimulate the diffusion and adoption of social innovations among followers and late adopters.
f. Develop a comprehensive set of indicators regarding social innovation in the public sector, that can be used by policymakers and academics in future studies.
g. Explore future trends of public sector innovation based on the outcomes of the research project as well as through interactions with academic and public sector practitioner expert groups.
In order to learn from the cross-sectoral and cross-country comparison of drivers and barriers we gathered data in eleven European countries: the Netherlands, Belgium, France, Germany, the United Kingdom, Denmark, Estonia, Italy, Spain, Romania and Slovakia. The proposed research was focused on seven policy sectors that are relevant to the EU: social security/welfare services, (mental) health services, sustainable public services, public utilities, (electronic) procurement, information society/ICT (teleworking), urban/rural planning for regeneration. Moreover, attention was given to the role of regional and local authorities.
By pursuing these goals, this project aimed to learn from innovation in public sector environments. Furthermore, it aimed to disseminate its knowledge to the public sector in the EU, so that policy makers, public managers, citizens and other stakeholders can ‘learn’ from this study.

The attached Figure 1 shows the overall conceptual set-up of the project. The distinctive work packages elaborated thematically and methodologically on aspects of this overall model.

1. Innovation environments. It is important to understand how social networks - in terms of the position that people (e.g. politicians, professional representatives) hold, the leadership role that they define as relevant for innovation and the type and quality of the relationships which they maintain - may contribute to innovation in government. As such, we recognize the importance of the environment component in innovation processes. Public sector innovation does not only depend on the actions, power and beliefs of one person. It is the environment which seems to be influential and it is environment which seems to generate the resources, capabilities and capacities that are needed to take on the innovation challenge. These aspects were primarily addressed in Work Package 1.
2. Innovation inputs refers to the wishes and needs that are expressed by relevant stakeholders that lead to exploring new ways of thinking and working, and new services in relation to social and economic challenges. An important challenge in social innovation is how to ensure that the knowledge, information and experience of relevant stakeholders is considered. This involves a process of co-creation and participation. However, the power of the stakeholders involved varies, as does the type of resources that they bring in. Despite this, there is little empirical knowledge regarding the way in which so-called ‘weak interests’ are organized in social innovation processes. The term ‘weak interests’ especially refers to the participation of citizens. Work Package 2 was focused on how citizen participation and the protection of so-called ‘weak interests’ can be organized in co-creation processes in order to generate services and outcomes that are truly responsive to their needs. However, not only the citizens themselves play a role in voicing their interests and needs. Intermediary organizations can have a role in this. For instance, accountability organizations and public complaint organizations, and the knowledge that they have, can be used as inputs to public sector innovation in order to give non- or weakly organized citizens and companies (as possible end-users) a voice. The complaints formulated by ombudsmen, and also the findings of independent audit offices and other social accountability practices, need to be taken into account. Work Package 3 have looked into the role of complaints and recommendations as possible inputs for public services innovation.
3. Innovation tools and processes. Alongside the inputs used in public sector innovation, the processes and tools used in social innovation are important. We distinguished three processes.
a. The first is the innovation process itself. This refers to the process in which new ideas are explored, developed, contested and implemented. As mentioned above, how relevant stakeholders and their inputs are used in this process is vital.
b. The second process refers to the diffusion of an innovation and the adoption of this innovation by other organizations, which also influences the impact of an innovation.
c. The third process is the learning process which takes places in and between the involved stakeholders in order to explore new ideas and services, and then to implement them. In order to support these processes, various tools are used. In this learning process, the definition and management of risk plays an important role. Innovation refers to the willingness and capacities to explore new ways of thinking and doing. In assessing a new idea, it is important to understand whether it is worth exploring or implementing, also considering the associated risks. In these learning processes, organizational, political and societal risks are taken into account. This is important given the varied interests, the different target groups and stakeholders (politicians, citizens, professionals) and values which are at stake. It is known that risk definition and selection play an important role in the pursuit of social innovation in the public sector. However, we did not know yet what factors and circumstances influence the social and political construction of risk. This is an important topic as the next step was to develop instruments which support public managers, senior civil managers and politicians in dealing with risks in a more systematic way. In order to develop this, empirical knowledge is needed about the various kinds of risks and their assessments. Work Package 4 was focused on risk definition and risk management. Another vital process is the diffusion, adoption and upscaling of innovations. The ultimate success of a social innovation depends on its adoption by other organizations and people. In addressing the impact or outcomes of an innovation, it is also important to look at its diffusion. As such, the diffusion process can be considered as a process that is located at the interface of two elements: processes and outcomes.
4. Innovation outcomes, diffusion and upscaling. The success and failure of social innovations in the public sector depends on the degree to which these innovations are adopted by other organizations and whether they have the desired outcomes. However, adoption research to date has primarily focused on the first ‘early adoption’ stage of the diffusion process, with hardly any attention having been given to the question of how to reach ‘late adopters’. This latter stage refers to the scaling-up process of diffusing an innovation. What is typical of this process? How can we develop instruments and tools which facilitate this scaling up process? What role do innovation champions play? Work Package 5, ‘Diffusion, adoption and upscaling’, therefore was focused on this issue.
5. Feedback loops in innovation systems. If innovation can be considered as a learning process, then it also important to understand how specific knowledge and information can be used in new innovation processes. This means that we had to pay attention to the utilization process of the acquired knowledge in this research project. A first way in doing so, was developing a set of indicators that help to determine the social innovation capacity of organizations, networks and regions. Here, we developed a set of indicators which takes into account the characteristics of the innovation environment and of different innovation processes in order to assess the innovation capacity of an organization, collaborative network or region. Work Package 6, ‘Indicators of innovation’, responded to this question. The development of these indicators also took into account emerging issues and trends in the social innovation environment of the public sector. Further, we needed to ensure that the knowledge developed in the proposed project is utilized by both public administration practitioners and scholars. This was done throughout Work Package 7, where we first pulled together the lessons derived from the other work packages. In this way, we translated the findings of this research project in a systematic and policy-relevant way. This package further analysed emerging new issues and trends in public sector innovation and translate them into future scenarios. A work package on disseminating the results to the wider public and public sector practitioners (Work Package 8), together with the project management (Work Package 9) completed the project.

Project Results:
In this section, we summarize the achievements of each individual work package within the LIPSE project, as well as the wider activities aimed at creating impact and disseminating findings to the wider academic and policy communities. We do so work package by work package, and present a summary, an overview of significant findings, and a description of how the work was managed and how it corresponds to the original objectives of the research. Obviously, this summary only presents a snapshot of each work package, and we refer to the academic articles for in-depth analysis and background.

Work package 1: Innovation capacities of Innovation environments

This Work package, led by Roskilde University, is examining the innovation capacity of four cities – Barcelona (ESADE), Copenhagen (Roskilde University), Rotterdam (Erasmus University) and West Lothian (adjacent to Edinburgh) (Edinburgh University). It looks especially at the role of trust/social capital and leadership as relevant drivers of innovation capacity.
The objectives of the Work package were a) to identify relevant drivers and barriers that explain the innovation capacities of these environments, b) to identify relevant drivers and barriers that explain the innovation capacities of these environments, c) to provide practical guidelines to policymakers on how social innovation structures can be created that exploit social capital and the type of leadership that is needed to stimulate this an d) to disseminate the research findings.

Significant results
In order to accomplish the set objectives of this work package, we conducted a Document analysis of the organizational structure of the municipalities, a survey of administrators and politicians, and interviews with community-based innovators were used to gather information on the innovation capacity of the municipalities (Copenhagen, Rotterdam, Barcelona and West Lothian - adjacent to Edinburgh).

This resulted in the following findings:
• Innovation capacity is related to the innovation environment. For municipalities, this includes national governance structures and societal traditions, the local socioeconomic context, and formal organizational structures. There are some striking similarities in the socioeconomic challenges faced by the municipalities (demographic changes, economic growth, unemployment, health care and educational problems). In regard to innovations, a mixture of internally driven innovations, concrete products, service delivery innovations and externally driven innovations were described. Each of the municipalities nominated different drivers for innovation. Copenhagen had the most positive view of its structures, procedures and context as supporting innovation.

• Municipalities rated their own innovativeness differently to what could be expected based on local socioeconomic conditions, international innovation rankings, and the perceptions of those outside the municipality. Copenhagen leads the international rankings (out of these four), and the community rating of its innovativeness, but has the lowest self-rating. The socioeconomic challenges and significant innovations nominated by the community were very similar to the nominations from within the municipality for Copenhagen and Rotterdam, but quite different for Barcelona.

• Informal networking activities are crucial to innovation, because they provide the means for overcoming the ‘hard wiring’ of formal structures, allowing people to meet in more open and informal spaces, inside and outside an organization. Openness is important for innovation, as are boundary spanners who can bridge the gaps between groups to improve information flows. Barcelona stands out as having the most external contact and doing the most boundary spanning.

• Network ties to generate new information and ideas, and closeness to support smooth exchanges, are both important for innovation. In each of the municipalities, the informal networks are shaped by the formal organizational structure. Brokers play an important function in innovation through their capacity to draw upon diverse sources of information. Two types of brokers were observed in each of the municipalities – one with diverse and unconnected ties, and one with ties that are linked to each other. Both types might be necessary in public sector environments.

• Finally, leadership is also important to innovation capacity. It influences individuals’ scope to put forward new ideas within an organization. A ‘motivator-risk taker’ leadership type is regarded as important in each municipality, which combines skills in motivation, collaboration and risk-taking

Resources and deviations
The following factors have caused delays and changes. First, Edinburgh municipality refused to participate so West Lothian was included instead. Secondly, negotiations with municipalities caused delays in the starting stage, especially in Edinburgh and Barcelona, Thirdly, Barcelona needed to remove the questions in the survey related to trust, in order to get the municipality’s agreement which took some time and which also delayed the beginning of interviews in stage 2. The willingness of potential respondents to be interviewed in stage 2 also generated delays. Fourthly, one key- researcher from Edinburgh was unable to work for 5 months due to illness. These issues have been discussed with the responsible project officer who allowed that the report and policy brief would both be presented in December 2014, thereby granting an extension of the project for two months.

Work package 2: Co-creation and citizen involvement in social innovation

This work package, led by Erasmus University, focuses on factors that shape the process and outcomes of co-creation in social innovation. It especially addresses the role that citizens play in social innovation processes. A qualitative, case study approach is being followed, where relevant stakeholders are interviewed about their views on co-creation and co-production with citizens. Case studies in two policy fields are carried out: social welfare/social security and urban/rural regeneration. Seven countries are involved: The Netherlands (Erasmus Univeristy), The UK (University of Edinburgh), Denmark (Roskilde University), Germany (Hertie School of Governance), Estonia (Tallinn University), Spain (ESADE) and Slovakia (University Mateja Bela).

The main research objectives were: a) to identify the different types of co-creation, thereby focusing on positions of the involved actors and the coordination mechanisms that are used, b) to identify and to compare relevant drivers and barriers that account for the success or failure of co-creation processes between EU countries , c) to assess the outcomes of social innovations that are based on co-creation in relation to the expected benefits for the involved stakeholders, including the weak interests of citizens, d) to make policy recommendations regarding instruments for co-creation arrangements and e) to disseminate the research results and policy recommendations among involved policy makers and within the academic community.

Significant results
The work package took off by conducting a systematic review to all the relevant literature on co-creation and co-production. This review led to the drafting of a theoretical framework, which formed the basis for our case studies. From February 2014 until August 2014, each partner in WP2 conducted two case-studies (one for social welfare and one for urban/rural regeneration) which have been selected from the long list. After drafting the initial research report, Erasmus University Rotterdam - together with Tallinn – organized a panel and international focus groups with relevant opinion leaders and involved stakeholders (deliverable 2.4). Here, we discussed the main findings regarding the major drivers and barriers and the outcomes of co-creation. By doing so we were able interpret and validate our research findings. Also, we discussed with possible policy recommendations needed to be considered.

This resulted in the following findings:
• The policy context played an important role in the success of co-creation and that it is not recommendable to simply copy ‘best practice’ examples from another country. Additional measures to adapt the project are always important. They are usually aimed at creating financial sustainability and/or improve relationships between stakeholders.

• We also uncovered several organizational factors that played an important role: 1) high levels of risk aversion among the city administration have negative consequences for co-creation. 2) Co-creation was raised smoothly in places where public officials had a positive attitude towards citizen engagement. In other cases, the belief among public officials that citizens are not skilled enough to shape or provide public services was a main obstacle. 3) Third, we found that lack of long-term financial support was often the biggest burden for the co-creation initiatives’ long-term survival. 4) Fourth, political attention could act as both a driver (i.e. gaining more notoriety for the initiative), but also as a barrier (i.e. prioritizing the wrong interests and making co-creation political accountable). 5) Fifth, the role of scale and scalability appeared to be important. This relationship is parabolic. On the one hand a limited scale hinders expansion possibilities. A too large a scale causes barriers in communication with other stakeholders.

• On the citizen side, we examined another four important factors: 1) first, we concluded that co-creation initiatives have a higher success rate if the citizens feel ownership over them and their development. 2) Social capital was an important accelerator of co-creation in every case and was usually strong even before the initiative started. 3) Third, citizens’ willingness to engage in co-creation despite its many challenges was absolutely essential to an initiative’s success. 4) Finally, gaining a reputation of being a reliable service partners helped the initiatives to proceed.

• Regarding the outcomes of co-creation, we reached the following conclusions: 1) co-creation initiatives are usually developed alongside traditional public services, not as a complete replacement of them. 2) It is still unclear whether co-creation leads to more effective and efficient public service delivery, especially since it is difficult to compare across co-creation projects. 3) In some cases, choosing citizen organized services instead of the traditionally offered ones may reduce public accountability. After all, citizens are not publicly elected to hold office and make public decisions. 4) On a related point, the co-creation initiatives we reviewed were usually led by a group of middle class, relatively well educated citizens, which although not negative in itself, does raise questions about the representativeness of these projects and their impact on equity in public service provision. 5) Our last and most unexpected finding was that co-creation is in itself a method for strengthening social capital in a community as it brings people together and encourages relationships to form between citizens.

We conclude that given the importance of social innovation in the coming years—and the potential upsides and downsides—embracing and further researching co-creation is a timely and very important endeavor for policy makers, managers, street-level bureaucrats and researchers alike. We should keep researching it, acknowledge its downsides and not see it as a ‘silver bullet’ for challenges contemporary member states are facing.

Resources and deviations
The research report for WP 2 was submitted on the 31th of January 2015, together with the policy brief. Due to some editing procedures, the research report was submitted one month later than agreed upon in the Description of Work. The responsible project officer Dr. Yuri Borgmann agreed on this change. The overall time-span of the work package was not exceeded, nor changes needed to be made in the budget.

Work package 3: Mapping and analyzing the recommendations of ombudsmen, audit offices and other emerging accountability practices as inputs for social innovation

In this Work package (led by Leuven University), we focus on reports and recommendations made by accountability mechanisms such as audit offices, ombudsmen, and other emerging arrangements. More specifically, we are interested in the contribution of these reports and recommendations to the anchoring of social innovation in the public sector. The intermediary, accountability and complaint arrangements to which we refer, have been established in order to give a voice to citizens and end-users, the so-called ‘weak interests’. The recommendations of these accountability institutions can be considered as relevant sources of input for innovation.

The main objectives of this work package were: a) To make an inventory and analysis of international and national databases on best EU practices for innovations relating to accountability practices and instruments, b) to analyze the recommendations made by audit offices and ombudsmen, as well as other accountability mechanisms, c) to identify relevant drivers and barriers that explain if and why these recommendations have (or have not) been translated into new products and new service delivery processes, d) to make policy recommendations in order to improve the use of accountability information for public service innovation and e) to disseminate the research results and policy recommendations.

Significant Results
In order to accomplish these objectives in work package 3, a number of national and international databases in which best practices are described were analyzed. The analysis of the best practice databases involves an online survey among the identified best practice cases. For this purpose, the Leuven team developed a survey questionnaire, which includes questions about the further life courses of the best practice cases (dependent), and questions about the explanatory factors, notably: feedback, accountability, learning. Further, in order to investigate the recommendations made by ombudsmen, audit offices and other institutions, an extensive literature review was conducted, generating a set of potential drivers and barriers that may explain if and why recommendations of ombudsmen/auditors have (not) been implemented. These include factors such as culture, power relationships, media attention and economic incentives. On the basis of this literature review, the Leuven team developed a protocol for the analysis of ombudsman and audit reports, a protocol for interviews with the ombudsmen/auditors, and a protocol for interviews with members of the organizations under scrutiny. After completing over 70 interviews and gathering data from almost 250 good practice cases, we then turned to the analysis of our data.
Based on these steps we have found the following results:
• Cases in our innovation database came predominantly from the public health sector, social welfare sector and general administration.
• Most of the innovations in our database focused on e-Government, quality assurance, efficient procedures and citizen involvement.
• The first causal mechanism: FAL → Z proved to be able to partly explain the sustainability of social innovation. Awarded innovations who have ceased to exist were in general characterized by a lower FAL-score than those who still existed.
• Through the qualitative research into the implementation of ombudsmen and SAI recommendations on the second causal mechanism ((FAL, X) → Z) it was found that the reports of Ombudsmen and SAIs significantly influence the sustainability of innovations
• The qualitative research also highlighted the importance of factors explaining social innovations independent of FAL or X, explaining the third causal mechanism: ((FAL, X) or Y) → Z.
• The organizations who correlated significantly with sustainable social innovations were characterized by:
o The concerns of staff, customers and ombudsman impacting strategic decisions
o A sense of responsibility amongst employees
o Transparency about their results towards external stakeholders.
o A culture of open debates, the encouragement of experimentation and a forgiving culture if and when these experiments would fail.

Resources and deviations
The research report for WP 3 was submitted on the 28th of March 2015, together with the policy brief. Due to some editing procedures, the research report was submitted one month later than agreed upon in the Description of Work. The responsible project officer Dr. Yuri Borgmann agreed on this change. The overall time-span of the work package was not exceeded.

Work package 4: Risk governance in social innovation

This Work package – led by Edinburgh University - is focusing on how risk definition and risk selection influence the shaping of social innovation practices in the fields of mental health and environmental sustainability. In this we aim to develop approaches which can support public managers, senior civil managers, politicians and citizens and service users in engaging with these risks in a more systematic way. This could help users to improve the governance of risk in innovation programmes.

Objectives of this work packager were: a) Empirically identify and explain to what extent it is possible for relevant stakeholders to engage in discussions about levels of risk for public service innovations and how these discussions are translated into specific risk management and governance models, b) to make recommendations regarding the formulation of relevant principles for effective risk governance in innovation in public services and c) to disseminate the research results and policy recommendations among involved policy makers and within the academic community.

Significant Results
Therefore this work package has conducted a review of relevant literature and conceptual framework. In order to align the ideas and activities of the participating researchers the working group has organized a starting meeting in March 2014. At the same time, preparation has taken place in order to design and pilot the survey. This survey is needed to quantitative examine the two selected policy sectors. It is targeted at 100 officials working with innovations for each policy sector in each country. The survey maps the extent and nature of the risks definitions identified in innovation processes, the current approaches of risk management that are used as well as how they are assessed. Subsequently, two case studies are selected - based on the survey - and conducted.

Most important findings include:
• Providers of public services in the area of sustainability tended to be small, young, and private non-profits, whereas mental health public service providers tended to be either based in the public sector or associated with the public sector through contracting or Public-Private Partnerships (PPPs). They also tended to be bigger and more established, although this did not necessarily have any consequences for the current state of risk management.

• Risk perceptions were almost entirely fuzzy. Few respondents felt comfortable to provide a definition for the term “risk” in the context of their organisation. However, when provided with examples or when free qualitative responses were analysed, there was clear evidence that every organisation necessarily followed at least one risk management approach, although in most cases not explicit based on the risks of social innovation.

• Generally, bigger and more established organisations tended to exhibit more professional risk management approaches in place. In the Netherlands, however, even bigger organisations did not show much awareness of risk management in social innovation outside of governmentally set standards.

• In the case of mental health, risks were more tightly controlled through common standards in the form of regulation or legislation. Sustainability, on the other hand, seemed to benefit from more flexibility, but also a more confusing legal and regulatory landscape as there was no single source of legislation or regulation in any of the four countries (unlike for mental health).

• It is important to note that service-level staff was also engaging in risk management albeit they did not a) identify their activities as such or b) pursue their risk management strategies in any systematic or interconnected way. Initiatives tended to emerge based on individual teams and in the guise of “making work easier/safer”. This seems to confirm the image of front-line staff as “street-level bureaucrats” that affect policy through their day-to-day activity. Further research is needed to explore how such unintentional policy-making affects the results of higher level policy-making.

• Overall, the risk discourse is dominated by actuarial risk approaches that focus on financial gain or loss above all other risks. This was attributed to a focus by most funders and governments on financial data to indicate success or failure. While some organisations reported that learning was part of the innovation process, and failure therefore inevitable, this atmosphere was limited to consequences within the organisation rather than affecting service users or funding relationships (such as contracts with the public sector or private funders).

• The risk discourse takes place mainly at the level of management with few designated risk managers across public and private organisations. This was attributed to a lack of funding for overhead staff to take on the role of risk managers. Moreover, the interviews also indicated a perceived public attitude that such roles were not part of the role of public service provider, in particular in the non-profit sector.

• Regarding the discourse on risk itself, the data strongly suggests that the predominance of actuarial risk management results in a negative connotation of risk. Rather than seeing it as a necessary part of any social innovation, risk is still perceived as a concept to be minimised or even avoided, an attitude, which comes at the cost of social innovation initiatives. For many organisations across all four countries, in particular those working closely with the public sector, risk was a perceived “no go”. Non-profits, in particular in sustainability, were dependent on funders’ willingness to tolerate the risks associated with innovation.

• The risk management and governance methods in practice showed little variation in technique, and differed mostly in the number of staff involved. Overall, a classical system of project teams reporting to project steering groups and boards, operating according to a detailed project plan, was the dominant form of risk management, if formal responses to risk existed. In most cases, interviewees also referred to formal and informal communication structures driven by project managers. As aforementioned, this system tended to be based on financial indicators and driven by managers, with little involvement of service staff. The described structure particularly applied to the mental health case studies. As a result, both innovation and risk management tended to be top-down, which is less risky than the bottom-up innovation strategy of the sustainability case studies.

• Innovation and risk management are primarily funder-driven in the case of sustainability, with more opportunities for bottom-up innovation and a (slightly) more pervasive awareness of risk management in social innovation. This can be attributed to the small organisation size and private non-profit nature of sustainability service providers. In fact, funding sources seemed to produce a discernable level of organisational isomorphism, i.e. it seemed to encourage similar structures across organisations whose main form of income was external and competitive funding. Targeting funders is thus an important part of affecting the risk and social innovation approach of public service organisations.

• The realisation of identified and unidentified risks in social innovation projects seemed to result in their general categorisation as a “failure” rather than an important learning opportunity to avoid future occurrences of the respective risks. This ties back to the negative and limited risk discourse within and across organisations and the public. The “blame game” (Hood, 2012) phenomenon seemed to dominate risk culture in so far as the avoidance of risk was seen as insulating the organisation (or individual employees) from reputational risks resulting from potential failure. Of course, such an outlook has consequences on the willingness to engage in social innovation, in particular when vulnerable service user groups are involved.

Resources and deviations
The research report for WP 4 is submitted on the 28th of March 2015. After a rewriting procedure, the policy brief is submitted on the 22th of September 2015. Due to some editing procedures, the research report is submitted one month later than agreed upon in the Description of Work. The responsible project officer Dr. Yuri Borgmann agreed on this change. The overall time-span of the work package is not exceeded.

Work package 5: Adoption, diffusion and upscaling of ICT-driven social innovations

Work-package 5 started in January 2014. It was focused on upscaling ICT-driven innovations in the public sector – in particular, e-procurement and telework – in relation to the specific characteristics of followers, late adopters and laggards. Six partners are involved: Bocconi University (WP-leader, Italy), Erasmus University Rotterdam (the Netherlands), ESADE (Spain), Mateja Bela University (Slovakia), the National School of Political Studies and Public Administration (Romania) and the Ecole Nationale d’Administration (France).

More specifically this work package aimed: a) to theoretically and empirically identify the drivers and barriers that play a role in upscaling ICT-driven innovations in two policy fields (e-procurement and telework), in relation to the specific characteristics of followers, late adopters and laggards in six European countries, b) to develop policy guidelines and instruments that public decision-makers can use to improve adoption, diffusion and upscaling of ICT-driven innovations and c) to disseminate the research results and policy recommendations.

Significant Results
Work package followed the following steps: First, the WP5 Theoretical Framework has been accomplished Second, a database on the ICT readiness of the six countries involved has been created. This has required the analysis of relevant website and datasets, such as those ones produced by the EU, the OECD and the UN. Third, the WP5 Qualitative Analysis has been launched. From September 2014 onwards, the qualitative analysis has been consequentially executed, with the aim of assessing determinants and barriers in the adoption and upscaling of e-procurement and telework. Furthermore a cross-country comparison has been carried out. Moreover, qualitative findings served as a basis for the WP5 Quantitative Analysis, which focuses on the determinants and barriers in upscaling e-procurement (at the regional level) and telework (at the local level). This implied the production of a questionnaire, the design of statistically relevant samples and the launch of the survey. From February 2015, WP5 activities were first of all focused towards the completion of the collection of survey data and the related analysis using a common statistical software: the postponement of the gathering of quantitative data had been authorised by the European Commission. Secondly, the comparison among the six countries was completed.

Based on these activities we have found the following findings:
• The research shows how particular determinants and barriers affect the adoption and upscaling of both e-procurement and telework. However, such factors are not strictly technical in nature. As human-executed processes, these ICT-driven social innovations require organizational changes that not only deal with ICT skills and capabilities, but also represent basic enabling conditions.

• In the case of e-procurement, legislative obligations and mimetic pressures are the most powerful drivers from the outer context, especially for later adopters and non-adopters. Conversely, regional budget constraints represent possible barriers to the implementation of e-procurement. Innovators are instead more heavily influenced by political factors, such as the support of politicians and the presence of inter-organizational conflicts. With respect to the inner context, robust managerial support provided by committed and visionary leaders is a strong facilitator for adoption and upscaling. Such actors can foster the implementation, enacting training and consulting activities which take into account both technical aspects of e-procurement and, more importantly, the organizational reshaping that this implies. Change management is therefore a tool to be employed for emphasising the benefits achievable through a fully-fledged upscaling of e-procurement and to overcome the obstacle of a risk-averse bureaucratic culture within organizations.

• The research on telework has highlighted the importance of need-based demands and geo-morphological contexts as drivers from the outer context. These elements are intertwined. The typical challenges faced by contemporary Western societies, such as the necessity of pursuing a better work/life balance, are also connected to geographical setting (e.g. densely inhabited urban areas). Public sector organizations are thus required to be responsive to such aspects, taking into account their territorial specificities. We also noticed how positive imitation can emerge as a powerful determinant among later adopters and non-adopters. With respect to the inner context, bottom-up spontaneous initiatives by employees can determine the success of telework adoption. These have to be supported by top management, in order to provide the necessary guidance and boundaries. A major obstacle is represented by a bureaucratic culture that focuses on processes rather than on results. In order to highlight the benefits achievable through telework (which are also economic in nature), training activities are fundamental, as they allow managers and employees to handle organizational and psychological spill overs of distance work. Finally, experimentation with telework on a narrow organizational basis is also positively influential, since pilot projects allow an organization to limit the risks of implementation and to convince more sceptical members of the organization.

Resources and deviations
The research report for WP 5 was submitted on the 30th of May 2015, together with the policy brief. Due to some editing procedures, the research report was submitted one month later than agreed upon in the Description of Work. The responsible project officer Dr. Yuri Borgmann agreed on this change. The overall time-span of the work package was not exceeded.

Work package 6: Public sector innovation indicators

In this work package (led by Tallinn University and Erasmus University) a new analytical framework for public sector innovation indicators was developed. This allowed us to develop a three dimensional model of public sector innovation indicators including the logic of impact (efficiency-legitimacy), logic of feedback (in-through) and logic of technology/ICT (leader-follower). In doing so, also the results of the work packages 1, 2, 3, 4 and 5 were included into the approach.
The goals of this Work package are a) to adjust and refine the existing overall framework based on an update literature review regarding the factor influence social innovation in the public sector, b) to identify social and public sector innovation types and indices, c) to identify major theoretical, methodological and practical strengths and weaknesses of existing sets of indicators and indices and d) to develop a theoretical and methodological framework for the development of social innovation indicators in the public sector as well as formulating specific indicators

Significant Results
Our research in work package 6 took place in four phases: state of the art review of public sector innovation; research of prior innovation measurement efforts (MEPIN, EPSIS, APSII, NESTA, GII); development of a new evaluative framework for the study of public sector innovation indicators; and lastly, the framework was actualized in two empirical studies (e-procurement case studies from the city of Tallinn and the global survey and analysis of new experimental spaces for innovation in the public sector). The first two phases of the research were completed and work on the framework and case studied was ongoing by the beginning of February 2015. The new analytical approach was finalised, in accordance with the DOW, after two international focus groups with practitioners and researchers at the end of April and beginning of May 2015 in Estonia. The approach was tested in an Estonian case-study on the procurement of public servceis. This was followed by the meta-analysis of the results of work package 1-5 (a separate LIPSE working paper was compiled to cover the results, also used as input for work package 7). Based on the findings the empirical analysis was completed by the end of August 2015.

Based on these activities work package 6 can draw the following conclusions:
• Next to productivity and performance dynamics, public sector innovation is, on the most abstract level, related to public sector authority and legitimacy. New technological developments and data sources (incl. social media, big data) create novel opportunities to capture the latter; as of now, however, these possibilities are essentially not utilized in public sector settings.
• Political demand for public sector innovation indicators is crucial in initiating, facilitating and funding measurement efforts.
• We proposed an evaluative framework for public sector innovations based on three basic logics: first, the logic of changes; second, the logic of feedback, and third, the logic of technology. This creates a new, three-dimensional evaluation framework for public sector innovation evaluation. This was discussed with various stakeholders, which led to further elaboration of the 3D evaluative framework.
• The cases from Tallinn illustrated that currently used evaluation frameworks are relatively narrow and often determined by limited public procurement frameworks.

Resources and deviations
No abbreviations or resource changes have occurred.

Work package 7: Future trends and scenarios

Work package 7 (led by Erasmus University) is aimed at conducting a validated analysis of relevant trends in the environment of public administration, combined with an analysis of the results of LIPSE, which has led to the development of four scenarios about the future of social innovation.

Significant Results
In order to draft these scenarios, the WP 7 project team conducted an analysis of the research results of the overall project. The review of the results as carried out in WP 6 was used as an important source of information. In doing so we were able to take stock of the relevant drivers and barriers as well as outcomes of social innovation in the public sector. Based on this aggregated overall view we have translated these results also in specific recommendations. Furthermore, the WP team completed her environmental trend analysis, based on information that was provided by the LIPSE members according a specific format that was used. Furthermore, additional reports were studied. Gazed on the TAIDA approach, four scenarios about the possible future of social innovation have been developed. These are validated by feedback given during the focus group (12 March, 2016 in Edinburgh) and during the three roadshows (Berlin, Tallin and Milan). As a result the scenarios were concluded and presented at the final LIPSE conference held on the 16th of June 2016.

Based on these analyses we conclude that given a number of large trends, which characterize the most urgent problems of the European public sector (e.g. ageing population, increased migration, climate change and feelings of insecurity), social innovation will come down to the willingness of involved actors to participate in processes of co-creation. Co-creation can then be defined as the corner stone of social innovation in the public sector in order to develop new approaches that are able to meet the needs of society and societal groups. The willingness of citizens and governments to participate can be defined as the most important but most uncertain factor. Based on this notion four possible scenarios can be drafted::
• High willingness of citizens and government → ‘Let’s dance’ scenario
• High willingness of citizens; low willingness of government → ‘Lone Ranger’ scenario
• Low willingness of citizens; high willingness of government → ‘Flogging a dead horse’ scenario
• Low willingness of citizens; low willingness of government → ‘Wasteland’ scenario

Scenario 1 (Let’s dance) is the most desirable scenario. Our research report elaborates on characteristics of each scenario and offers recommendations to how scenario 1 may be achieved if one feels himself stuck in another scenario.

Resources and deviations
WP 7 started one month earlier than initiatially planned. In doing so, the WP could more easily build forth on the other WP’s without any loss of time. The scenario study is delivered at the end of LIPSE (July 2016). This is three months later than initially planned, since we included feedback given during the IRSPM conference held in Hong Kong in April 2016. The responsible project officer dr. Yuri Borgmann agreed on these changes.

Work package 8: Dissemination

See next section

Work package 9: Project Management

The management of the project went relatively smoothly. No major incidents have occurred. Reallocation of resources were made in accordance with the project officer. LIPSE has organized several board meetings. The starting conference took place on May 7 & 8 2013. All partners were presented here. The second was organized at September 11 2014 at the EGPA conference in Speyer Germany. The then responsible project officer was also present at this meeting. Additional meetings were organized to discuss the progress of the project. These were held on: February 3 2015 (midterm conference Brussels and attended by our advisory board and project officer Yuri Borgmann); March 31 2015 (IRSPM conference Birmingham); August 27 2015 (EGPA conference Toulouse) and the last June 16 2016 (LIPSE final conference, attended to by our advisory board). Next to these meetings, meetings were organized within the scope of the different work packages and between work packages. Also the management team of LIPSE gathered almost weekly (see deliverable 9.8) to discuss the progress and urgent matters of the project.
In terms of personnel, many junior researchers joined the team during LIPSE, to assist and conduct the empirical work of LIPSE.

Potential Impact:
In the Description of Work of LIPSE, it was agreed upon that research and dissemination activities are geared primarily at:
- Impact on practitioners, policy makers and third sector
- Impact on the academic community
- Generic impact, aimed at all stakeholder groups

Impact on practitioners, policy makers and third sector

The LIPSE project is receiving increasing attention from policy circles. This is evident from a
large number of presentations and practitioner publications. Researchers have been invited to numerous public sector practitioner fora to present their work, among others from the municipality of Copenhagen, the OECD and the Dutch Ministry of Internal Affairs.

Impact on the academic community
Academic publications. The research results of the different work packages have resulted in numerous publications. These are published in academic journals, as edited volumes, special issues and books on public sector innovation (see overview in electronic participant portal).
Survey data and other datasets. The output of the various work packages, ranging from WP 1-5 form a unique set of comparative data, varying from survey data, interviews, document analysis and detailed case studies. This data is used to draw all kinds of comparisons across countries. As such LIPSE formed a unique opportunity to examine different aspects of innovation in a similar way in countries across the entire European Union.

Generic impact
Web-based dissemination activities. LIPSE has initiated a website containing projects outputs, such as research reports and policy briefs. Also it includes a link to our blog where LIPSE researchers have contributed to. Further the website links to a compilation video of what social innovation looks like in practice and what kind of lessons can be drawn from that. Last, the website offers a link to social media. In doing so, the results of LIPSE are accessible for everyone who is interested and is thereby reaching a much larger audience.

Main dissemination activities
While the full list of outputs form the LIPSE project is available through the electronic participant portal, we will below outline some of the main outputs and dissemination activities.

Working papers
To facilitate fast dissemination of findings prior to the publication of articles of chapters, key
findings have been made available through a LIPSE working paper series. A total of six working papers have been published.

Development of the Project Website
Since the initiation of the website in February, the LIPSE website ( had approximately 10.000 unique visitors and over 60.000 page views. Visitors mainly came from within the European Union: Netherlands, Germany, UK, Belgium, Italy, Denmark, Spain and Estonia. Outside the EU, visitors came from the US and Canada. Visitors were mainly interested in general information about LIPSE and the publications.

Media and social media
LIPSE has disseminated findings through media and social media where appropriate. A press release was issued at the beginning of the project. Besides the LIPSE account on Twitter (@EULIPSE), consortium members, including Lykke Margot Ricard, Lars Tummers and Hanna de Vries, use Twitter to communicate findings (@LGTummers; @Hanna_de_Vries;@LykkeRicard ). LIPSE and events have attracted various retweets and mentions on twitter (see for some recent results). We also launched a blog which can accessed throughout the website The Facebook page continues to attract followers - It now has 157 likes (a substantial increase compared to the second midterm report), and is used to post new reports and conference calls. In addition, most LIPSE researchers are active on Facebook and use their own feeds to communicate about new reports and articles. Various journals and professional magazines have also devoted articles to LIPSE findings. Examples include Public Administration and Public Management Review.
Series of Policy Briefs
Policy briefs have been written to summarize key findings of the projects, The policy briefs are written for policy makers, and provide easily accessible data and analysis, and are generally quite short. They are available both in printed and electronic versions in English, German and French. The policy briefs proved to be very useful for dissemination during conferences and events targeting policy makers. The following policy briefs have been produced:

1. Policy Brief WP1: December 1 2014
2. Policy Brief WP 2: January 26 2015
3. Policy Brief WP 3: March 28 2015
4. Policy Brief WP 4: September 22 2015
5. Policy Brief WP 5: May 30 2015
6. Policy Brief WP 6: December 31 2015
7. Policy Brief WP 7: June 29 2016

Research Reports
LIPSE has produced a set of research reports. These reports are either formal deliverables of the project, or are core outputs associated to one of the work packages. Below is a list of the research reports, sorted by work package:

1. Lewis et al. (2014) Work package 1: Innovation environments and innovation capacity in the public sector
2. Voorberg et al. (2015) Work package 2: Co-creation and citizen involvement in social innovation: A comparative case study across 7 EU-countries
3. Van Acker et al. (2015) Work package 3: Mapping and analyzing the recommendations of ombudsmen, audit offices and emerging accountability mechanisms
4. Flemig et al. (2015) Work package 4: Risk definition and risk governance in social innovation processes: A comparative case study across 4 EU-countries
5. Nasi et al. (2015) Work package 5: Determinants and barriers of adoption, diffusion and upscaling of ICT---driven social innovation in the public sector: A comparative study across 6 EU countries
6. Kattel et al. (2015) Work package 6: Public sector innovation indicators: Towards a new evaluative framework
7. Bekkers (2016) Work package 7: Social innovation in the public sector: drivers, trends and scenarios

LIPSE high level practitioners conference
A High level practitioner conference was organised in February 2014 in order to communicate findings to public sector reform practitioners. The event took place in the Academy Palace in Brussels, in close proximity to the European institutions. Speakers included both LIPSE participants, and representatives from various national ministries and ombudsmen, members of the advisory board and our project officer Yuri Borgmann.

LIPSE final conference
Near the formal end of the LIPSE project, we organised a concluding conference in Brussels (Belgium) on June 16 2016. During this conference, the main conclusions of our research on social innovation as well as prospects for future research have been discussed. The LIPSE-network invited a selected number of academic and professional experts to join us in this debate. The conference, which attracted approximately 50 academics and practitioners (European commission and agencies, national administrations and social innovators), consisted of a number of plenary sessions, round tables, and workshops. During the conference, the results of LIPSE were discussed, and confronted with the experiences of practitioners, working in social innovation practices.

Other conferences targeting public sector practitioners
Apart from these conferences, LIPSE was also present at a number of events with a more practical focus, rather than an academic focus. For instance, Prof. Bekkers gave a presentation at the OECD March 18 2014 and Prof. Kattel attended to an OECD expert workshop on February 4 2014. Also the Hertie school of governance organized a workshop for policy makers and public managers on April 25 2013 in Berlin. Other events involved the TAD dialogue in Lugano, attended by Prof. Lewis and Dr. Ricard and a presentation of the main results of WP 1 in Copenhagen on January 28 2015 also attended to by Dr. Ricard. And of course the roadshows, organized in Milano, Tallinn and Berlin organized in spring 2016.

Academic conferences
LIPSE was presented at multiple academic conferences over the last years, whereby the results of the research have been presented and discussed. This involved the IRSPM conference in 2013, 2014, 2015 and 2016. Also the EGPA conference was attended to in 2013, 2014 and 2015.

LIPSE also (co) hosted a number of academic conferences. These involved among others, Proceedings for the 2013 international scientific conference on ‘The EU budget, innovation and economic growth’ organized by Marta Orviská in Banská Bystrica (Slovakia), 5th November 2013. A seminar for the Melbourne school of government on February 9 2015, organized by Prof. Klijn and Prof. Lewis. Also on July 7 2016, the KU Leuven organized a seminar concerning the results of LIPSE in WP 3.

Collaboration with external partners and institutions
LIPSE is collaborating with all kinds of partners, both academic and professional organizations. For instance, Dr. Ricard has strongly collaborated with the municipality of Copenhagen who were very interested in the results of our research. The same goes for our cases within work package 2. Examples of co-creation in Germany, the Netherlands and Estonia were also interested in our research, resulting in a visit from the initiators to our final conference. Further, LIPSE collaborated with other academic networks and universities. For instance the School of International affairs and public affairs in Fudan, Shanghai China, hosted a consortium of partners including ASPA, IRSPM, PSA and LIPSE and was co-chaired by Prof. Osborne. Dr. Lars Tummers was member of the academic advisory committee. Furthermore, LIPSE partners were actively involved in the organization of various key conferences, such as EGPA (co-chair Permanent Study Group Rainer Kattell, varous members participated) and IRSPM (co-chair study groups by among else Taco Brandsen, Victor Bekkers and Lars Tummers, various members participated). Also, the National University of Public Service in Hungary hosted an event whereby LIPSE was also an organizing party. Last, LIPSE sought actively connections with other FP7 projects, such as TEPSIE, INNOSERV and WILCO, by visiting their conferences and to build forth on their findings important for social innovation.

Academic publications
Academic publications are listed in the main table of publications in the participant portal. A focus on articles in peer-reviewed journals has been at the core of the LIPSE academic dissemination strategy. This includes both individual papers and special issues. Special issues have been published or are currently under review with: Public Management review and International review of Administrative Sciences.

Open access strategy
In line with the ‘open access’ movement, and in order to ensure immediate and unrestricted access to research material, all reports have been made available at the project’s website. In accordance with the open access pilot project of the European Commission, launched in August 2008, and clause 39 of the grant agreement, project partners undertake to deposit peer reviewed research articles or final manuscripts resulting from LIPSE research into their institutional or if unavailable a subject-based repository.

Exploitation of results
The project has not resulted in directly commercially exploitable or patentable results. Instead, findings have been made available free of charge for various public authorities and other researchers. Exploitation of results is also happening though providing findings to various organisations. This continued interest reconfirms the attractiveness of the LIPSE project for policy makers and research institutions. Furthermore, the results of the LIPSE project will be used to draft a new proposal for the CULT COOP H2020 call 11, in which the topic of co-creation in social innovation will be translated in relation to the topic of developing and testing new forms of collaborative service arrangements. Some of partners of LIPSE will also participate in this new call. LIPSE results are also used in the BELSPO project that is coordinated by the University of Antwerp and in which LIPSE researchers participate. This project, funded by the Flemish government started in 2016.

List of Websites:
The website can be accessed using the following address: