by| Pierre Rossel | Cristiano Codagnone |Gianluca Misuraca
Policy resistance in an age of complexity
Today, society and the economy are more interconnected, unstable, and unpredictable than ever. Furthermore,developments in Information and Communication Technologies (ICT) are happening at a very fast pace.
The Internet as we know it today is already a remarkable catalyst for creativity, collaboration and innovation, providing possibilities that would have been impossible to imagine just two decades ago. If one had predicted then that, today, children would freely access satellite images of any place on earth, interact with people from everywhere and search trillions of data with a simple click on their PCs, one would have been taken for fool [Misuraca et al., 2010].
Current policy making strategies and also the ways of procuring supporting evidence for policy decision making are no longer able to cope with complex, multidimensional and highly dynamic societal challenges. For more than 60 years, society has largely failed to eradicate critical social challenges despite investing increasing resources into state policy resistance is responsible for these failures. Policy resistance occurs when an intended policy outcome is defeated intentionally or unintentionally by complex and dynamic elements, agents, factors, first order and second order feedback loops, and so on. The causes are typically multidimensional and found throughout history [Sterman, 2006].
On the other side, we now have at our disposal a radical increase in computing power along with outstandingly widespread distribution of networked communities. The possibility of collecting and processing huge amounts of data at moderate costs was unthinkable only a decade ago. These developments have led to the emergence of futuristic visions, such as ‘singularity’ which suggests that computers will exceed human cognitive capabilities and an ‘intelligence explosion’ which could, among other things, prolog and improve quality of life [Kurzweil, 2005]. However, current tools and approaches for policy design, implementation and evaluation are ill-suited to capturing this complex and interconnected future. Moreover, they are based on an abstract and unrealistic vision of the human being: rational (utility maximising), average (not heterogeneous), atomized (not connected), wise (thinking long-term), often highly simplified (complexity denial) and politically committed [Piniewski, Codagnone and Osimo, 2011 forthcoming].
In short, the intellectual framework upon which policy making rests is no longer adequate. Our claim is therefore that a paradigmatic shift in developing a new policy modelling framework is required. However, this is not simply a matter of more computing power and more data. Multiple longstanding challenges may also need to be addressed.
Critics however may wonder if the inherent complexity of our free living systems increased. Is today’s society and economy more unstable and unpredictable than ever? Certainly changes are ubiquitous. The world is increasingly interconnected via the Internet and other new media. Most would agree that complexity and unpredictability were also robust thirty or forty years ago. As a matter of fact, since 1957 the neoclassical homo oeconomicus approach (that choices are made based on fully rational human) (see Herbert Simon, 1957), has been challenged. In 1985 the American sociologist Marc Granovetter [1985], advancing beyond his first 1973 classic on the ‘strength of weak ties’ [Granovetter 1973], argued that rationality is socially embedded and not exercised in a social vacuum. In other words, we act according to the socially shaped structure of opportunity we face, using and being influenced by the network of social relations into which we are embedded.
These important social pressures may go unobserved and not be considered in policy design, implementation and evaluation thus enabling a mounting systemic policy resistance and defeating the policies using only the traditional tools of policy making such as regulation and incentives3. In other words, when changing tastes and preferences are influenced by social interaction, a simple ‘stick and carrot’ incentive based policy or regulation may be impotent in effecting desired change. In addition, restricting full freedom of choice may backfire and trigger unintentional systemic resistance to the policy. However, opportunities for policy-making may be enhanced precisely because we do not act strictly according to instrumental rationality. Our actions are not always self-interested yet may still lead to contributions that can be harnessed to achieve policy goals. Citizens can be unobtrusively and intelligently helped to make optimal choices supporting both their individual well-being as well as group well-being policy goals. In addition to this, a number of technological, economic, societal, political and environmental trends and developments affect all countries as well as most policy domains. In order to deal with the challenges associated to these developments a new culture of future oriented thinking is needed [Havas et al,2010].
Our claim is thus that combining foresight and ICT-enabled modelling techniques in support of governance and policy-making may be useful to improve policy intelligence. More specifically, embedding foresight methodologies in policy modelling techniques may lead us to a new generation of policy making, so to avoid the often shortsighted and piecemeal approach of current decision-making that is usually incremental and step-by-step, and does not pay sufficient attention to changes in the environment and cross-policy dimensions.
ICT for Governance and Policy Modelling: a possible solution?
Given the significance of globalisation, increasing technological and organizational changes as well as the even increasing importance of learning capabilities and applications of knowledge, our future cannot be predicted by any sophisticated model in a sufficient and reliable way [Havas et al, 2010]. As for policy-making itself, there is a widening gap between the speed, complexity and uncertainty of technological and socio-economic changes, on the one hand, and the ability to devise appropriate policies, on the other. Even the credibility of science is somewhat fading. Scientific research no longer stands for ‘true’ in itself and the ‘objectiveness’ of policies based on models is questioned as scientists and ‘modellers’ themselves are known to have different opinions and models often come to different conclusions on the same issues.
Within this evolving context, the European Commission launched in 2009 a new area of research on ICT for governance and policy modelling. According to the European Commission’s 7th Framework Programme (Work Programme ICT 2009- 2010) [European Commission, DG Research, 2009a], ICT for governance and policy modelling joins two complementary research fields, which have traditionally been separate: the governance and participation toolbox which includes technologies such as mass conversation and collaboration tools; and the policy modelling domain which includes forecasting, agent-based modelling, simulation and visualisation. These ICT tools for governance and policy modelling aim to improve public decision-making in a complex age, enable policy-making and governance to become more effective and more intelligent, and accelerate the learning path embedded in the overall policy cycle [European Commission, DG INFSO – 2008a].
Within this framework, in 2010 the European Commission funded a Support Action to design the Future Research Roadmap on this domain: CROSSROAD -A Participative Roadmap for ICT research on Electronic Governance and Policy Modelling, aiming at defining a shared vision, able to inspire collaborative, interdisciplinary and multi-stakeholders research. CROSSROAD in fact links very diverse research disciplines with practitioners’ views and policy makers’ concerns, through a multi-stakeholder and participatory approach and provides an useful tool for the support and orientation of future policy-making.
Overall, the CROSSROAD research roadmap aims to push to new outreach options the boundaries of traditional eGovernment research and help resolve the complex societal challenges Europe is facing by applying ICT-enabled innovations and collaborative policy modelling approaches, which include the harnessing of collective intelligence, agent-based modelling, visual analytics and simulation, just to mention a few [CROSSROAD, 2010a, b]. In this context, CROSSROAD aimed at building a consensus-driven Research Roadmap to consolidate and advance research in a new, yet highly fragmented, domain and to provide strategic directions for the future of research in ICT for governance and policy modelling. The main goal of the CROSSROAD project has been to drive the identification of emerging technologies, new governance models and novel application scenarios in the field of governance and policy modelling, leading to the structuring of a beyond the state-of-the-art research agenda, fully embraced by research and practice communities. In summary, CROSSROAD identified and characterized the key research challenges in the field of ICT for governance and policy modelling and ultimately outlined a concrete, participative roadmap for future research.
Conceptual and Methodological Framework
Conceptual frame and Objectives
The reasons for developing forward looking analysis to support policy decisions stem primarily from the emergence of important science and technology applications and their wider implications for society. Science and technology interact with society in a complex way and their ‘effects’ are often neither immediate nor direct, but of second or third order and occur after a substantial time delay [EC, JRC-IPTS, ESTO, 2001]. More specifically, technological developments in the domain of governance and policy modelling happen at a fast pace. Policy-makers cannot afford to wait until situations are clarified and until the effects are evident before they take decisions. Though tomorrow’s developments are uncertain they originate in conditions established today. Hence, there is an important need for policymakers to scope the impacts of science and technology and how they may develop [Da Costa et al. 2003].
The history of forward-looking analysis and future studies spans decades [EC-JRC-IPTS, FOREN, 2001] and three main areas of future-oriented technology analysis can be identified [Cahill and Scapolo, 1999]: technology forecasting analyses the conditions and potential of technological development within a concrete framework; technology assessment supports decision making by generating technology or problem-specific options arising from new developments; and technology foresight addresses the impacts of technological development on a broader scale. However useful these methods may be, the growing knowledge-intensity, the pace of technological and societal change and the increasingly distributed and networked character of the economy and of governance processes cannot be explored using only technology-oriented future studies [Compano and Pascu, 2005]. A more comprehensive approach is required. Designing scenarios relies on foresight methods, which are based on a much broader concept than technology assessment and forecasting. It calls upon a wide range of themes and stakeholder perspectives, in order to examine the social and economic aspects of future technological developments. The process is interactive, open-ended and bottom-up and paves the way to identifying possible breakthroughs and exploring implications and hypotheses that will support the definition of strategic directions and policy related decision-making [EC-JRC-IPTS, 2003c].
The objective of this paper is therefore to present and discuss the main findings of the scenarios for Digital Europe 2030 designed by IPTS as part of the CROSSROAD’s project and based on a foresight exercise which included: 1) an analysis of the key areas of expected change in the domain of ICT-enabled governance and policy making to be placed in the context of various different future scenarios, and 2) envisioning, for each scenario, the risks and opportunities offered by ICT tools for governance and policy modelling techniques, as regards their contribution to overall EU policy goals. Based on these findings, aimed to explore possible alternative futures in governance and policy making, the paper elaborates further exploring new research frontiers embedding foresight methodologies in the future expected mainstreaming of participatory ICT tools and policy modelling techniques.
Methodological approach
This paper is partly based on the results of the research carried out by CROSSROAD, an FP7 Support Action to design the Future Research Roadmap in the domain of ICT for governance and policy modelling. This Action aimed to provide strategic directions and define a shared vision, able to inspire collaborative, interdisciplinary and multistakeholder research. In this context, a participatory foresight exercise has been conducted outlining a set of scenarios on how governance and policy modelling, supported and enhanced by the use of ICT, could develop by 2030 in order to identify the research needs and policy challenges to be addressed. To design such scenarios, an analysis of future needs, risks and opportunities under different conditions was conducted based on the current state of the art of the domain [CROSSROAD, 2010a]. The scenario design exercise resulted in four different scenarios which explore how governance and policy making could develop by 2030. The scenarios were developed by means of narration (storytelling) of possible future outcomes in selected key domains of European society where the development of ICT tools for governance and policy modelling techniques are likely to have a major impact. By looking at the future of ICTenabled governance through four thoughtprovoking visionary scenarios, the research helps policymakers to foresee what European society could become twenty years from now, thanks to advances in ICT for governance and policy modelling. The scenarios, their formulation and interpretation, expose the gaps that exist today in research and what needs to be addressed in order to enable better governance and construct a more open, innovative and inclusive digital Europe tomorrow.
Scenarios in fact are systematic visions of future possibilities. In foresight research, this usually means plausible possibilities that do not rely on extreme wild cards. [Miles, 2003]. They are used as tools for political or strategic decision making and to explore the future impact of particular decisions or developments [Janseen et al, 2007]. More specifically, Scenario building aims to identify uncertain developments in the future and include them as elements of the scenario narrative [Janseen et al, 2007].
However, this exercise’s time horizon (i.e. 2030) and the interrelationships between different developments affecting it (e.g. rapid developments in specific domains of ICT) make the future of this research area dynamic, complex and uncertain, with little available evidence that can be used to predict or forecast these futures. Therefore the method of scenario design has been used for this exercise and it followed a common 5-step methodology: 1) a trend analysis to determine the developments that could be key drivers for the future of ICT tools for governance and policy modelling techniques, 2) the selection of the scenarios by determining the main impact dimensions and key uncertainties, 3) writing of the scenarios, 4), identification of the implications of the scenarios by participants at the Experts’ Workshop and by consulting the public and 5) deriving conclusions for policy implications and research challenges [EC, JRC-IPTS, FOR-LEARN,2010].
With regard to the methodological approach informing the scenarios design exercise, it must be underlined that foresight research comprises many different methods that can be categorised in several ways. According to Popper [Popper, 2008], these methods can be: expert-based, creativity based, interaction-based or evidence-based. The objectives of a foresight exercise and the degree of uncertainty and complexity involved guide the selection of methods for a particular exercise.
The Scenario Design adopted as part of the CROSSROAD’s foresight exercise is: 1) evidence-based, as it builds on the trends emerging from a policy review and trends analysis and the assessment of the state-ofthe-art of research in ICT for governance and policy modelling; 2) expertise-based, as it includes the views of experts gathered through a call for expert contributions and further discussion held during an Expert Workshop to validate the scenario design framework; 3) interactive, as it incorporates the inputs from the Expert Workshop and from online public consultation; and 4) creative, as it is based on the ‘creative-thinking’ that came out of a series of brainstorming activities by members of the IPTS lead team and other experts.
In this paper we attempt to go further exploring how combining foresight and modelling techniques in a dynamic and participatory manner may enhance policy intelligence capabilities and thus lead to a better policy making process. Foresight in fact can help in picking up ‘weak signals’: weak but very important hints that a fundamental re-assessment and realignment of current policies based on traditional models are needed. In other words, our hypothesis is that embedding foresight techniques in structured modelling platforms may serve as a crucial part of an early warning system, and it can be used as an instrument for developing policy intelligence mechanisms. In furtherance to this, participative, transparent, forward-looking methods may be instrumental to support policy-makers in finding solutions for complex societal challenges that cannot be addressed by traditional policy recipes and models based on evidence from the past. A dynamic component is needed to be introduced in modelling policies so to take into consideration appropriate changes and trends developments, as well as inputs from all interested stakeholders that would guarantee effective policy implementation.
Results and Policy implications
The impact of foresight on ICT for governance and policy modelling
The scenario design developed as part of the CROSSROAD’s research aims to provide a structured framework for analysis of current and future challenges related to research on ICT tools for governance and policy modelling techniques [CROSSROAD, 2010b]. The scenario framework proposed was chosen to stimulate further debate and reflection on possible, radical alternative scenarios and is, of course, one of several possible alternatives. It takes today’s world and constructs images of possible future worlds, highlighting ways in which key uncertainties could develop. The aim is to present clues and key impact dimensions, thus increasing the ability to foresee possible development paths for the application of ICT tools for governance and policy modelling techniques. Thus risks can be anticipated and better preparation can be made to take advantage of future opportunities. In turn, this outlines key elements to be taken into consideration for the further roadmapping and impact assessment of future research in this domain.
Instead of attempting to forecast several future ICT-enabled scenarios, it was chosen to define four internally consistent – but radical – views of what the future European Information Society might look like in 2030.
These give four distinctly different views of what Europe’s governance and policy making system could be and what the implications of each could be for citizens, business and public services.
The pace at which the elements of the visions unfold will, however, be influencedby the speed of change of the overall technological landscape and societal context. Considering the unprecedented growth and speed of ICT uptake in several research themes and the rapid emergence of technologies which enable applications for ICT for governance and policy making (e.g. social computing, mobile technologies, pervasive computing, etc.), we can argue that the world in 2030 will be radically different from the world we live in today.
Following the mapping and analysis of the state of the art in the research themes related to ICT for governance and policy modelling and the identification of emerging trends, the main impacts on future research in this area were defined. These were further refined through an analysis of existing scenario exercises and the current shaping of policies and strategies for the development of the European Information Society. They were then used to develop the visionary scenarios framework to depict possible state of the future Digital Europe.
The uncertainties underlying the scenario design are: 1) the societal value system we will be living in (more inclusive, open and transparent or exclusive, fractured and restrictive), and 2) the response (partial or complete, proactive or reactive) to the acquisition and integration of policy intelligence techniques in support of data processing, modelling, visualization and simulation for evidence-based policy making.
Accordingly, the key impact dimensions were classified on two axes: Degree of Openness and Transparency (Axis Y) and Degree of Integration in Policy Intelligence (Axis X) and they go from the extreme 0 (Low Openness and Transparency and Low Integration in Policy Intelligence to 1 (High Openness and Transparency and High Integration in Policy Intelligence). The axes represent ways in which social and policy trends could develop. Based on this framework scenarios were then developed in a narrative manner (i.e. storytelling style) as descriptions of possible outcomes in selected key areas, representative of the European context, where emerging trends related to the development of ICT tools for governance and policy modelling techniques could have an impact. The scenarios for the future of governance and policy making of Digital Europe 2030 are presented in Figure 1 .
The vertical axis indicates the degree of openness and transparency in a society, in terms of democratic and collaborative governance that could be further enabled by ICTs. The most open and transparent society would be one where even traditional state functions are completely replaced by non-state actors, through opening-up and linking public sector information for re-use. Such a society would be characterized by open standards and principles of transparency and accountability in governance and public management [Misuraca, 2009b].
The openness paradigm is also expected to apply to the research and business community which could benefit from open innovation and social/business networks of collaboration, where users are co-creators of products and services delivered globally via peer-topeer social networks based on reputation and trust [EC-JRC-IPTS, 2009a]. An important aspect will be the regulatory and technological solutions, and also the sociocultural attitudes to the basic digital rights underpinning the future Information Society. In fact, the concept of openness is not strictly related to technological solutions, but rather to socio-cultural and organisational aspects that can be enabled and supported by technological advancement [Misuraca, 2009a].
The horizontal axis concerns Integration in Policy Intelligence, i.e. the degree of integration of data and knowledge and way in which collaboration between all stakeholders in policy-design and decisionmaking mechanisms is enabled. This involves the possibility (enabled by ICTs) to mash-up data and information available from different sources in an ‘intelligent way’ (meaning efficient, effective and able to generate public value). It also involves the extent to which users, individually or as members of formal and informal social networks, can contribute to the co-design of policies, simulating and visualizing the effects of legal and policy decisions, and engage in real-time monitoring and prior assessment of possible expected impacts at local, regional, national and panEuropean scale. This axis is also associated with the capacity and willingness of policy actors to share power and change decisionmaking mechanisms in order to facilitate the redefinition of basic democratic freedoms in a collaborative fashion. This could go to the extreme of redesigning the traditional mission of the State and the role played by governance stakeholders. Again, ICTs are not the driving force; rather change is driven by changes in social values, attitudes and new paradigm shifts in terms of information management, knowledge sharing (experts vs. non expert networks, for example) and allocation of resources [Rossel, Glassey and Misuraca, 2009].
In all the scenarios, the world in 2030 is expected to be radically different from today’s, due to the unprecedented growth and speed of ICT uptake in several fields and the related impact ICT tools which enable governance and policy modelling techniques may have. Moreover, the influences and drivers of innovation and renewal in the public sector, combined with increased financial pressure on states will result not only in change, but will also affect the pace at which the state adapts to the new environment, to its new roles and to increased engagement with stakeholders and users.
However, whichever scenario dominates in the future, in the coming years, conventional wisdom and familiar governance models will be challenged as ICT-based disruptions impinge on democratic, consultative and policy-making processes. Evidence already gathered anticipates that the scope and scale of transformation will have a major impact on society [Broster, 2007]. Since 2005 there has been a phenomenal growth in mass, on-line collaborative applications, which has captured the imagination and creative potential of millions of participants – particularly the younger generations. In addition to new forms of leisure pursuits, communitybuilding activities have also entered the political arena as witnessed in a number of recent national elections [EC, JRC-IPTS, 2009a, 2009c, 2009d].
Online communities can leverage considerable human knowledge and expertise and rapidly build their capacity. At the same time, it is now recognised that online collaborations have the potential to trigger and shape significant changes in the way future societies will function. Extrapolation of the present exponential growth leads to scenarios where a very large percentage of the population could, if equipped with the appropriate ICT tools and capacities, simultaneously voice opinions and views on major and minor societal challenges [Tapscott and Williams, 2006]. Hence, these tools herald the transition to a different form of dynamically participative governance models.
While such scenarios are readily imaginable, we also recognise that we currently do not have appropriate governance models, process flows, or analytical tools with which to properly understand, interpret, visualise and harness the forces that could be unleashed. Present government processes (local, regional, national and EU level) provide laws and regulations, interpret and define societal norms and deliver societal support services. Their legitimacy is derived through democratic processes combined with a requirement for transparency and accountability. In a world that is increasingly using non-physical communication and borderless interaction, the traditional roles and responsibilities of public administrations will be subject to considerable change and classical boundaries between citizens and their governments will become increasingly blurred [Pew Internet, 2010b]. The balance of power between governments, societal actors and the population will have to adapt to these challenging new possibilities.
A key issue will therefore be to develop and apply advanced ICT tools to provide robust support to the change process and facilitate the transition to a new digitally derived legitimacy. Inherent in this process is the definition and realisation of new, carefully crafted governance models. By 2030, there will no longer be any barriers which prevent citizens and businesses from participating in decision making at all levels, and hence the present democratic deficit will be overcome.
Advanced tools – possibly building on gaming and virtual reality technologies will enable citizens to track the totality of decision making processes and see how their contributions have been (or are being) taken into account. Current linguistic and cultural barriers will have been largely overcome through use of semantic-based cooperation platforms [Broster, 2007]. Opinion mining, visualisation and modelling into virtual reality-based outcomes and scenarios will help to both shape, guide and form public opinion. These ICT-enabled processes and tools will have to demonstrate transparency, earn people’s trust and be devoid of manipulation. The outcomes of such consultative processes should be faster and more efficient policy revision and decision making.
By 2030, it is expected that transparency and trust in governance processes and policy making will characterise a changed relationship between governments, businesses and citizens. Governments traditionally collect, process and store significant quantities of data. In the future, the relationships will change and businesses and citizens will be in a position to ‘authorise’ access by governments to ‘data spaces’ of their own data which they control and update. Such a scenario would result in a ‘private shared space’ jointly accessed by data users and data providers [Reutter, 2008]. Equivalent data spaces will be adopted by businesses. These shared spaces will require extremely robust access rules and procedures and hence new technologies and ICT tools that ensure privacy and data protection. Trust in such technologies will need to be earned [EC, DG-INFSO, 2009].
In most organizations, information travels along familiar routes. Proprietary information is lodged in databases and analyzed in reports and then rises up the management chain. Information also originates externally: gathered from public sources, harvested from the Internet, or purchased from information suppliers. But the predictable pathways of information are changing: the physical world itself is becoming a type of information system [EC-JRC-IPTS, 2003b]. In addition to this, more objects are becoming embedded with sensors and gaining the ability to communicate. The resulting information networks promise to create new opportunities, improve governance processes, and reduce the costs and risks of policy decisions. In what is called the Internet of Things, sensors and actuators embedded in physical objects -from roadways to pacemakers- are linked through wired and wireless networks, often using the same Internet Protocol (IP) that connects the Internet. These networks churn out huge volumes of data that flow to computers for analysis. When objects can both sense the environment and communicate, they become tools for understanding complexity and responding to it swiftly. What is revolutionary in all this is that these physical information systems are now beginning to be deployed, and some of them even work largely without human intervention [EC, 2009e].
The widespread adoption of the Internet of Things will take time, but the time line is advancing thanks to improvements in underlying technologies. Advances in wireless networking technology and the greater standardization of communications protocols make it possible to collect data from these sensors almost anywhere, any time. Ever smaller silicon chips for this purpose are gaining new capabilities, while costs, following the pattern of Moore’s Law, are falling. Massive increases in storage and computing power, some of it available via cloud computing, make number crunching possible on a very large scale and at decreasing cost [Chui et al., 2010].
Research in the area of the Internet of Things is now strictly linked to advances in the field of Ubiquitous Networks and pervasive computing. Future applications are opening up huge opportunities for private and public sector organizations alike. Despite the fact that many of the technologies which underpin the future Internet infrastructure are not new (e.g. Radio Frequency Identification, sensor networks, GRPRS, UMTS-HSDPA and Near Field Communication, to mention a few), the conditions for their application may result in innovative and disruptive usages on a daily basis in forthcoming years [Pew Internet, 2010a]. This innovation could support several public policies, such as logistics, security, transport, environment and energy, education and health, and many others [Medaglia, Chicca, Orlando, 2010].
Implications of integrated foresight and modelling in support of governance and policy making
The scenarios developed as part of CROSSROAD served as an input to be compared with the integrated analysis of the state of the art in the domain of ICT for governance and policy modelling and, based on this comparison, a gap analysis has been conducted to identify an exhaustive list of specific gaps, where the on-going research activities are not going to meet the long-term needs outlined by the future scenarios. This exercise resulted in a substantial contribution to shaping the roadmapping of future research in the domain thus proving to be useful and needed. Through a participatory foresight processes it was possible to bring together not only experts and interested parties from academia and research, industry and government, but also involve directly policy-makers and other interested stakeholders. The documents produced in fact were made available online for comments and feedback and received a general appreciation during discussions at specific workshops and conferences.
This also demonstrated that with such an open and participatory approach in mind, it is now increasingly being recognised that an opening of the political process is required to ensure robustness and effectiveness of its outcomes. In recent years in fact we have assisted to a shift in policy making practices from shaping framework conditions and structural settings towards strategic decision making. However, the growing complexity of governance and policy making processes is also recognised.
A shift towards evidence-based / model based policy making is happening, but this is sometime not supported by effective empirical data and conceptually sound understanding of the societal implications of modelling techniques per-se. As a matter of fact, this shift in policy-making is also reflected in the evolving practices and interest in modelling techniques worldwide and in the EU in particular (see for example the Climate Change debate or Energy and Transport policy developments). However, in spite of its apparent success, initial enthusiasm is already given way to a significant deal of scepticism, both from ‘traditional modellers’ and non-experts, including policy-makers themselves.
More recently, and this is consistent with the results of the CROSSROAD roadmapping exercise, as well as the policy direction the EU is focusing on, it has been recognised that the effectiveness of policy depends also on the involvement of a broader range of stakeholders than those formally in charge of policy decisions.
This concept of distributed policy-making and intelligence originally set out by [Kuhlman, 2001] is de facto at the core of the foresight and roadmapping exercise underpinning CROSSROAD, where it is assumed that openness of governance systems and integration of policy intelligence can harness collective intelligence, building on the knowledge, experience, and competence of various actors. Applying this network perspective to a ‘distributed platform’ based on ICT enabled policy modelling and integrated foresight techniques (appropriately supported by participative and user-friendly simulation and visualisation tools), may prove to be instrumental to further implement policies and achieve socio-economic impacts, generating a ‘cascade’ of public and private decision-making on society’s course of change and affecting the interactions that precede formal policy-making processes.
In addition to this, behavioural change may also be stimulated as participating in the governance and policy making process may also enhance effectiveness of policy implementation from individual users and stakeholders other than the government. The role of government in fact is shifting from being a central steering entity to that of a moderator of collective decision-making processes.
However, in order to perform this role effectively, all stakeholders should be able to contribute to the policy directions commonly agreed, and governments need to be capable of setting up a shared platform for policy intelligence, where foresight and modelling techniques –if actually supported by ICT- can be crucial for improving governance and policy making processes.
Conclusions
Policy challenges and possible solutions
The scenarios developed aimed to define how the advancement of ICT tools for governance and the integration of policy modelling techniques could affect governance and policy making twenty years from now, so as to identify what research is needed and which policies should be promoted. Indeed, challenges in the emerging domain of ICT for governance and policy modelling are huge and complex and cannot be dealt with in isolation. In this regard, there is also a strict relationship with the broader task of envisioning and developing the Future Internet. The Internet was not originally designed to serve massive scale applications with guaranteed quality of service and security [Zittrain, 2008]. Emerging technologies like streaming high quality video and running 3D applications, or, in our specific domain, applications that enable mass collaboration, data processing, simulation and visualization through complex modelling, face severe constraints as regards running seamlessly anytime, anywhere, with good quality services.
European scientists have proved they are at the forefront of ICT research since the invention of the web and throughout the rapid technological developments of the last 20 years [EC ISTAG, 2009]. It is now time to bring together different research disciplines that could help us benefit from the opportunities of ICT for better governance and policy making, and at the same time overcome the possible risks to society of mainstreaming large scale applications in this domain. Additionally, and from a technological infrastructure perspective, we should remember that the current Internet, as a ubiquitous and universal means for communication and computation, despite being extraordinarily successful so far, has a series of inherent unresolved problems. It is expected that it will soon reach its limits as regards both architectural capability and capacity [EC, 2009e]. However, the future development of Internet infrastructure will be supported by complementary advancements in technological applications that are now consolidated trends and expected to grow even faster. The groundwork in place for years now should yield innovation in the near future [Pew Internet, 2010a]. More powerful devices, even cheaper netbooks, virtualization and cloud computing (including portable solutions), reputation systems for social networking and mass collaboration tools, as well as the proliferation of sensors, reporting and decision-support systems, do-it-yourself embedded systems, robots, sophisticated algorithms for processing data and performing statistical simulation and analysis, visualization tools to report results of these analysis, affective technologies, personalized and location-aware services, facial and voice recognition systems, electronic paper, anomaly-based security monitoring, self-heating systems and others are expected to become reality and mainstream in the next 10-20 years.
But far more important than network requirements and technological applications is the consideration of socio-economic aspects in the development of future ICT tools for governance and policy modelling techniques. Socio-economics as a multidisciplinary field, which cuts across all research areas [EC DG Research, 2009b] of the ICT for governance and policy modelling domain, has manifold research challenges. Suitable governance and policy-making mechanisms, which provide appropriate incentives for participation, but at the same time ensure security and avoid risks (of enlarging digital exclusion, for example), need to be designed.
Moreover, legal and regulatory issues such as digital rights, privacy and data protection, also have to be taken into consideration, as the demand for the establishment of trust in governance may increase (or shift) as its usage scenarios change [Hildebrandt, 2009]. For example, an ever-increasing openness of ICT enabled governance and policy modelling mechanisms, and the criticality and value of the transactions conducted over the open platform used for this purpose, may create incentives for malicious use of data and information. While security technologies will be developed to address the technological challenges linked to this, additional risks to trust arise, mainly due to its potential pervasiveness, large scale and involvement of users. The challenges include, for instance, the design of identity management systems capable of dealing with billions of entities, and their different roles in the governance sphere, the trustworthiness and control of distributed applications based on services offered through open service delivery platforms, and the secure and trusted interaction with real-world objects and entities through sensors and actuator network infrastructures [Pew Internet, 2010a]. More specifically, for example, the emergence of wireless networks could allow software applications and physical objects to be connected, opening up a wide range of stimulating new application scenarios in governance and policy making [Feijoo et al, 2009; Jaokar and Gatti, 2009]. At the same time, however, the same openness underpinning their mass-development and usage will expose sensor networks and related information and content to possible attack and misuse.
The opportunities provided by future ICT tools for governance and policy modelling for individuals, businesses and governments are huge but they will only be taken if appropriate conditions and ‘governance models’ are developed. In fact, it is expected that ICT tools for governance and policy modelling techniques will force change in institutions, no matter how resistant they are. And even if it could be predicted that governments that redefine their relationship with their stakeholders will be the ones to succeed, the market will still drive that process in the commercial domain, and tensions may emerge as stakeholders know more and more about the organizations that are trying to serve them [Pew Internet, 2010b].
At the same time, it seems that increasing demand from the scientific and business community, and from civil society organizations and citizens groups, will drive the emergence of ‘experimentally-driven research’, to address broad governance and policy-making challenges, developing and applying ICT tools and applications to exploit the full value of the mass collaboration and open and participatory paradigm underpinning the future technological developments and governance directions in Europe. This would eventually allow the testing of new ICT-based solutions and models for collaborative governance and participatory policy modelling, and socio-economic impact assessment of future societal changes. This last issue entails building on the momentum that the domain of ICT for governance and policy modelling has recently gained, by further developing the research community.
In order to bridge the gap between various stakeholders and long-term research and large-scale experimentation, enabling cross-fertilization across different scientific disciplines and integration of resources, special emphasis should be put on fostering common research results. This will create value for the EU, avoiding fragmentation of research efforts and it should also include the experiences gained at the international level. This requires developing a joint strategic research agenda, on ICT for governance and policy making to support the building of an open, innovative and inclusive Digital Europe 2030.
Future Research and next frontiers
Most remarkable and perhaps not comparable with the development of the Internet in its first evolution (what can be defined as the Web1.0) is also the exponential growth of the new generation of Web2.0 applications, both in terms of the number of applications and number of users. Remarkable too is the lightning speed with which the trend spread. It took barely three years to social computing to grow from a marginal community and become the dominant Internet trend which it is today.
The fast growth and massive uptake of Web 2.0 services are at the origin of a deeper socio-economic impact, the signs of which are however not clear yet. In fact, despite the rise of Web2.0 applications and its fast growth and pervasiveness, it is still quite difficult to capture the phenomenon and ‘measure it’ or even just building an empirically sound case for assessing specific impacts and its potential policy-relevance. Evidence of impacts of Web2.0 on our society is largely anecdotal and in most cases not systematically gathered and analyzed.
Existing metrics are not able to make sense of the transformations enabled by these emerging technologies as the changes they convey seem to be more behavioural and cultural than primarily ICT-driven. As a matter of fact, we are already witnessing several changes in our daily lives, and in personal and professional attitudes, especially if we look at the way the young people integrate their digital and real selves, or at how social networks and user-generated content is used and consumed (if not abused) (Misuraca, 2011 forthcoming).
In foresight terms, the momentum that has characterised the Web2.0 phenomenon is expected to continue, to further evolve and to mature. The driving forces and added values of it in fact reside in the practices (the values of social engagement) rather than in specific technologies and their sheer corresponding numbers. In the coming decade, Internet access and network bandwidths will continue to increase, and the Web, either as we know it today, or in yet surprising evolutions, will undoubtedly continue contribute to the development of the Information Society, shaping new form of user participations.
In this regard, the measurement/modelling issue becomes crucial, particularly in the context of informing evidence-based policy decisions. The most urgent need is certainly for new metrics to address the emergence of new social media, and in general, for systematic measurements and internationally comparable data. These would enable better assessment of the long-term importance of Web 2.0 trends in terms of their socio-economic impact, and the quantitative and qualitative differences between countries across the world. This is especially necessary in order to bridge the gap between the wealth of “marketing-type” data and the lack of official statistics, which occurs for every new socio-techno-economic trend, especially in the fast-evolving ICT landscape (Misuraca 2011 forthcoming).
At the same time, despite not yet supported by consolidated evidence, it seems that the most promising user-enabling ICT applications are emerging in the area of mass-collaboration for governance and policy-making, where mobilisation of politics and civic engagement is already in some cases producing a shift in the power balance between the “crowd” and political representatives. Moreover, web2.0 applications and values can support gathering collective intelligence of citizens and framing public opinion formation on specific policy-relevant issues in a structured manner so as to harness evidence-based policy-making and improve quality of regulatory and policy frameworks.
However, while ICT-supported modelling techniques are largely available to support impact analysis in specific policy areas, they often remain stand-alone models built in isolation and as ‘black-boxes’. They can contribute to respond to focused economic and techno-economic questions (e.g. impact of regulations on specific energy emissions or transport mode shifting) but cannot provide a comprehensive analysis of complex cross-sectoral issues that would affect the overall society (CROSSROAD, 2010).
In this regard, attempts to develop more sophisticated and integrated ICT-enabled models able to capture the various variables and consequences affecting societal changes are underway. However, in both the Web2.0 realm and more in general the ICT for governance and policy modelling domain, it should be considered that the quality of input that can be gathered from users through user-enabling technologies (e.g. social computing, mobile technologies, sensors, etc.) is highly variable, and filtering this content is still very much a resource intensive task. One key challenge is therefore to make sense of gigantic quantities of qualitative data, such as entailing mass conversations. In other words, the goal is to look at ways to improve signal-to-noise ratio, through a variety of means, with different human-computer balance, through tools such as sense-making, reputation management and collaborative filtering. In addition, visual analytics and simulations techniques (for example using virtual worlds or serious gaming) can help ‘domesticate’ and generalise results of modelling techniques to the wider public (CROSSROAD, 2010).
In conclusion, our assumption is that as current modelling techniques are not really adequate to predict, monitor and evaluate policy developments and their impacts on society, a new policy-measurement paradigm is required [Misuraca and Rossel, 2011, forthcoming]. As a consequence of what precedes, and broadly of our own research journey, we need to emphasize the relative absence of a meta-level of analysis, a reflexive layer where not only we would model, but also measure and model how we measure and model and with what implications. This epistemic concern is mainly about modelling issues. Basically, we could say that so far, obsessed with the need to play along the success story-line of more ICT the more welfare, we have downplayed the necessity to model our observations and discuss our modelling assumptions so as to improve them and more generally generate a knowledge process.
This broad perspective could for example build on the potential that gathering collective intelligence combined with advanced ICT-enabled policy modelling techniques. A shift is thus required, not only by enabling users to become ‘living sensors’ and providing data to be directly fed into comprehensive models, but also giving the possibility to the same users (being they researchers, businesses, civil servants or citizens) to have direct access to data they need, and process them using ICT-enabled simulation and visualization ‘intelligent’ systems (i.e. able to find meaning in confusion, independently of human-acquired knowledge). Ultimately this will not only allow to have a better measurement of policies (thus modelling and assessing the real implications of policies) but will also create new opportunities for people to interact with and influence governance and policy-making processes and make progress in solving societal problems; and therefore start to establish the target not only in the offer/use and therefore empowering paradigm as it has been profiled so far, but on more ambitious impacts and transformational options.
While so far this has not been the focus of research in the foresight field, we also argue that it would be required to explore this ‘foresight-modelling couple’ so to better grasp the potential of ICT dynamics, and especially user-enabling technologies for modelling and evaluating policy-options. This will also allow changing the perspective of the observer, thus gaining insightful evidence and data directly from the users, and in real time, and more generally speaking of usefulness/relevance of the accomplishments of ICT applications “in situation of use”, towards the great variety of ways of addressing small and grand challenges.
In this regard, as suggested by Piniewski, B., Codagnone, C, and Osimo, D. (2011, forthcoming), it may also be worth exploring the popular and innovative approach to crowd management strategies presented in the best selling book Nudge [Thaler, 2008]. In this book, the authors contrast the stylized agents of classical economics called Econs to more human-like agents called Humans. The Econ reliably reacts using his reflective cognitive system, whereas the Human is frequently unreliable in his reaction secondary to an automatic cognitive system. The fundamental game changer presented by Nudge is that traditional simulation efforts that depend on Econs will under-perform dramatically today as Human behaviours are significant contributors to crowd outcomes. Folks will spend money every year paying for magazines that are not read only because they fail to pay attention to an automatic renewal or malignant nudge. The authors claim that left to their own devices, Humans unlike Econs will often continue to make poor decisions affecting their own wellbeing. Humans are especially vulnerable when mapping (view to the future) outcomes is delayed or unclear at least for the moment. In this connection, linking foresight with modelling techniques may provide a better understanding of the issues at stake while also provide alternative policy options to address societal challenges.
However, to effectively integrate foresight and policy modelling techniques it would be required to develop a more refined or high definition understanding of socially embedded desires, tastes, preferences, and behaviours of the policy recipients they would like to affect. In addition, an understanding of how networks are born, grow and develop over time would be important. The effects of policies do not occur in a vacuum. They occur within a social network. Thus nudging alone will not be sufficient. Nudging plus network approaches raises challenges but also creates tremendous opportunities for innovative policy making [Omerod, 2010]. In a system of interconnected agents, changes by a few agents may produce a cascade of changes in many agents as they learn from each other, copy each other, and seek each other’s acceptance. If the network is scale-free (characterised by a power law) then changes enacted by the hubs can even more likely lead to a cascade effect. Such cascade effects may drift a policy into unknown or unexpected directions. Thus by understanding the basic structure and flow of a network, a small nudge can be applied to relevant hubs to trial the cascade effect at a smaller scale. Once confident that desired policy outcomes are reliably attained, a larger nudge can be applied to the network with the intention to scale effective policy. This can be repeated in cycles enabling continuous improvement co-production of effective community strategies. For instance, integrating information from classical surveys into agent-based modelling or other innovative ICT supporting modelling and simulation tools, policy makers would have access to a reasonable understanding of a network structure.
However, for such a paradigmatic shift in policy making to happen, it would be required that a policy intelligence platform enabling meaningful participation of multiple stakeholders will be realised. Citizens at large must have an effective voice to inform policy, to better understand choices affecting their futures and to take personal ownership of the actions that affect their daily life and future developments. This will provide an opportunity to vastly improve the evidence base upon which the policy cycles depend (design, implementation, evaluation).
Theoretically, a community of scientists and experts would produce the most comprehensive and robust form of evidence support for policy-makers. However, this insight may be delayed, difficult to use or not directly useful for specific policies. In turn, citizen-generated data may not be fully comprehensive (self-selected, biased, opinion) or robust (in need of filtering/ validation). Sensitive issues such as security and privacy must also be addressed. For these reasons, citizen-generated data are slow to enter the policy making arena.
The paradigm shift in policy modelling will occur when the top right quadrant of the matrix characterizes the bulk of policy evidence activity. In a cycle of continuous improvement, collaboration across the three key groups of stakeholders (policy-makers, expert scientists and non-expert citizens) will drive evidence-based policy. Together these stakeholders will support innovative data intensive policy action, capable of timely reaction and redirection within networked systems.
In summary, ICT alone cannot solve everything and can be even generate new problems. For the desired paradigm shift to occur, both institutional and cultural changes are needed. Figure 3 is a visual summary of the elements needed for evidence-based effective policy making. The key elements needed for ICT to advance as a powerful instrument across the value chain of data collection, data analysis, and support for action have been identified by the FP7 project CROSSROAD which reviewed a large body of literature to generate a roadmap for ICT use in policy modelling.
The figure provides a synoptic visualisation of how various agents (policy makers, citizens, scientists and expert) share and collaboratively use data through distributed computing. Inside the ICT tool box are tools for data analysis, data presentation, and for persuasive feed-back. Together, all stakeholders are able to obtain answers to their queries and collectively optimize policy to optimize citizen behaviours.
Within this framework, it is clear that data and information are a fundamental building block upon which the paradigm-shift in policy modelling depends. However different data may come in different formats and be difficult to link correctly. More importantly, data about the future are not available, and even the more sophisticated model will not allow to predict exactly what impacts specific policies may have, due not only to ‘wild cards’ events, but especially because of the not rational neither linear evolution of policy directions.
Therefore, it is important to complement current modelling approaches with participatory foresight techniques, allowing for example the possibility to gather data and opinions directly from users. In this manner, strategic and critical information can be volunteered by users and the dramatic reduction in cost of consumer electronics is increasingly making sensor-based devices less expensive and more popular. Such sensors enable both participatory sensing (requiring a minimum level of efforts from users) and opportunistic sensing (no user effort required, users to simply carry around smart devices that spontaneously collect and transmit relevant data). All of this can be integrated with the ambient data collected by Internet of Things (IoT) sensors and other devices capturing important information (i.e. satellite images).
This will allow developing innovative policy intelligence platforms that based on advanced participation, new modelling and simulation techniques which will take advantage of applications pushing the boundaries of analytics and visualisation techniques to help bridge the knowledge asymmetry between the experts, the policy makers and the citizen. Foresight will become a crucial component as the real time dynamic of such a policy intelligence platform will not rely simply on data about past and present facts, but will provide the framework for alternative policy options and related impacts, and thus possibly anticipating the future.
In this context, highly specialized knowledge and analysis will become more accessible while retaining the robustness of rigorous analysis. Static visual analytics will advance to interactive visualization with supporting analytical reasoning and scenario-design to help make well-informed dynamic decisions in changing complex situations. Problems once unknowable due to their size and complexity may become quite knowable. Combining foresight with policy modelling and especially visualisation techniques will provide a new set of tools extending from the presentation of discussion arguments in argumentation map formats for minimizing the complexity of policy debates to the creation of virtual environments which can simulate the behaviour of both policy makers and citizens in a real-life like environment. Such techniques will be the building blocks of new integrated policy intelligence platforms. In this regard, the recent call of the FP7 ICT WP 2011-2012 reinforced the focus of the research in the area of ICT for governance and policy modelling and intends to further advance the understanding of how emerging ICT tools for governance and policy modelling can provide opportunities for decision-making in a complex world, through the dramatic and combined growth of data available, analysis and simulation tools, participative and behavioural change technologies. Integrating foresight methodologies in this process is required and represent in oyr modest view the next frontier to be overcome.
Acknowledgements
This paper is in part based on research conducted by the Information Society Unit of the JRC IPTS within the CROSSROAD Project – A Participative Roadmap on future research on ICT for governance and policy modelling, an EU FP7 Support Action conducted by the consortium composed of: National Technical University of Athens, Tech4i2 Ltd, University of Koblenz Landau, EPMA and the European Commission’s Joint Research Centre IPTS (www.crossroad-eu.net).
