Designing the exercise
- Setting the objectives
- Determining the coverage (focus and scope)
- Defining the users
- Setting the time horizon
- Costs and funding
- Information gathering
- Project team and organisation of the exercise
- Communication strategy
- Designing the methodology (methodological framework)
Running the Exercise
Follow-up of the Exercise
The 'Foresight for Transport' project was supported by the European Community under the 'Competitive and Sustainable Growth' Programme (1998-2002). The project was undertaken in order to test the applicability of Foresight methodology for "visioning" transport and mobility futures and specifying impact pathways. The implementation of the study entailed organising thematic expert panel consultations (on the topics of enlargement, environment and energy, multilevel governance, information and communication technologies and time dynamics), a Delphi survey involving 165 experts around Europe as well as the establishment of a meta-database system with information on indicators that can be used to monitor developments in fields of relevance for transport and mobility.
The decision to set up and run a Foresight exercise in the form of a strategic dialogue targeting (European) transport policy is linked to the recognition of the limitations of mainstream policy assessment methods and, more specifically, transport models, with regard to the identification and analysis of the impact of non-transport factors on transport and mobility.
The majority of existing transport models operate with rather straightforward, albeit simple, assumptions. Thus all or most external and policy changes are 'transformed' into changes in either travel time and/or transport costs and these variables are subsequently used to estimate outcomes. The spatial assignment of the latter is done with reference to the characteristics of the infrastructure network and background characteristics, such as population density or past trends.
There is, in principle, nothing wrong with basing complex computations on simple assumptions. Simple assumptions are good not least for being easy to understand and communicate. If they are right they lead to correct projections. If they are wrong, the projections will be wrong and so the assumptions will need to be corrected. A more serious problem has to do with the so-called ecological fallacy. The ecological fallacy consists of deducing equivalent associations at the micro-level from relationships or associations we observe on the aggregate level. The correlation between a high quality infrastructure network and the degree of congestion does not necessarily imply, for instance, that investment in infrastructure will reduce congestion. That this occurs will equally depend on the timing of the investment, the mode of infrastructure and its location as well as the availability of alternatives. Many an investment made in the hope of reducing congestion in particular areas has not had more than a very short-term effect. Policy which is solely based on projections from transport models which operate with associations observed on the aggregate level is therefore likely to be misguided.
Designing the exercise
Starting out from considerations such as those alluded to above, the 'Foresight for Transport' study had as its main objective the task of clarifying the pathways through which external and/or policy variables impact on transport and mobility. Our research was motivated by two key considerations: first, to provide a framework for thinking through policy interventions while at the same time paying attention to the interrelationships between relevant factors; and second, to establish an analytical basis which is complementary to transport modelling and which, hopefully, can also be used to advanced the theory and practice of the latter.
Geographically the exercise focused on the European Union. The 'Foresight for Transport' project focused on the following external dimensions:
- Social Policy
- Science & Technology
- Environmental factors
- Economic developments
- Political and institutional arrangements
The classificatory scheme of external dimensions was identified during the first round of the expert panel consultations organised by the project. The latter were organised around five themes: (1) environment and transport; (2) ICT and transport; (3) the challenge of enlargement for transport; (4) time dynamics and transport; (5) governance arrangements and transport.
The 'Foresight for Transport' study was designed to provide decision support to transport policy-makers at different levels of government and as a complement to the outputs of transport models.
The time horizon for the 'Foresight for Transport' exercise was the period from the present until after 2020, distinguishing between the short-term (2004-2009), medium-term (2010-2019) and long-term (2020+).
The study's expenditures totaled just under a million euros. About a third of the budget was spent on organising consultations and included the running of workshops, the commissioning of external expertise as well as the fieldwork for the Delphi survey. The sponsor of the research was the European Commission.
The 'Foresight for Transport' project is unique in its coverage of several fields in addition to its focus on transport and mobility. This follows from its principal objective to study the impact of non-transport factors on mobility and the transport system.
The project was launched following a thorough assessment of previous work on transport policy, on one side, and foresight methodology, on the other. The first step of the project involved organising extensive reviews of the non-transport fields under investigation. The first round of expert consultations assisted in deepening this knowledge.
Information was collected by the core team and during the first round of the expert consultations. The Delphi survey was used to validate the conclusions of the expert consultation workshops regarding the link between transport and non-transport factors as well as scenarios for the future. Based on the results of the first round of the Delphi survey, together with additional research, the core team elaborated impact pathways and submitted this to validation at the second round of the Delphi survey. A second expert workshop refined the results and assisted in their interpretation with reference to transport policy.
The 'Foresight for Transport' project was carried out by a consortium involving 'The Interdisciplinary Centre for Comparative Research in the Social Sciences' (ICCR) as the coordinator, in collaboration with Adelphi Research (Germany), the University of Cardiff, School of Social Sciences (UK) and NESTEAR (France). The study was implemented over a period of just over two years and engaged a total of 16 scientists for different periods of time. By reason of its scope, the team built up for the 'Foresight for Transport' project was interdisciplinary, involving transport policy analysts, political scientists and sociologists. Specialised knowledge on specific fields was obtained by organising expert panels.
The impact pathways that link external elements to mobility and the transport system were produced in a four-step process:
- First, we gathered an expert knowledge base on each of the external dimensions. This involved understanding what drives developments for each, what the main contextual issues involved are and how these are likely to play out in the future.
- Second, measurable indicators were defined for each dimension and available data was collected in an attempt to describe actual trends and estimate how these might develop in the future. This established a basis for the reference scenario, i.e. that scenario which best describes the present and the latter's trajectory into the future.
- Third, we envisioned alternative ways in which the various dimensions correlate with each other and with transport and mobility. Such alternative global futures provide the setting for specific impact pathways.
- The fourth step was that of specifying the impact pathways at the micro-level as well as the degree of association between factors that are linked along this pathway.
A variety of techniques can be used in Foresight, including expert consultations, which, in turn, can use brainstorming and scenario-writing analysis. The Delphi survey method is also often used to draw more experts in the consultation process and also to encourage a convergence of opinion or allow convergence patterns to be identified. Quantitative methods like trend extrapolation or simulation modelling are used when trend data is available. The 'Foresight for Transport' study used all of these techniques, thus combining both qualitative and quantitative methods.
The emphasis on process leads to the participatory approach which is at the core of a Foresight exercise. This in turn prescribes the use of creative/consultation methods such as expert consultations, brainstorming and scenario-writing. These methods allow specialist information to be gathered and integrated, as well as validating it through reflection. A further advantage of Foresight is that it raises awareness among participants regarding transport policy and how it is related to other policy domains.
Important preconditions for these methods to also meet their objectives are: (a) the setting up of a core group of researchers or moderators who can follow the various deliberations on a continuous basis, filter and integrate their outputs; (b) a generous time framework.
As far as the time framework is concerned it should be underlined that the two years of the 'Foresight for Transport' project can at best be thought suitable for a pilot exercise but should not be taken as a benchmark for a strategic dialogue. This is in part because future estimations are long-term and must be revisited at regular intervals. An equally important reason has to do with the time that is necessary for deliberations as such, especially when these happen in part virtually or at a distance. A Delphi survey, for instance, can only provide reliable results if it achieves a reasonable response rate and this is not easy to achieve within a limited time framework.
The 'Foresight for Transport' project organised two rounds of expert consultations. The first round took place during the project's first year (June 2002), lasted a full week and was used to gather knowledge relating to external dimensions. These expert consultations were organised around five themes (environment, governance, communication technologies, enlargement and time politics) and involved around fifty experts. The second round of expert consultations took place towards the end of the project in October 2003 and lasted three days. This was a much smaller consultation exercise involving twelve experts. The objective of this consultation was to receive feedback to the project's findings on the transport impact pathways and the monitoring system.
The expert consultations, and especially the first round, used the brainstorming and scenario-writing techniques to guide the deliberations. To elicit ideas, the thematic expert consultations organised by the 'Foresight for Transport' project were structured around the following question: What are the important drivers and/or key issues with regard to [enlargement; energy and environment; ICT; governance; time dynamics …]? Subsequently, and in order to help cluster the various drivers and issues, participants were asked to think in terms of their temporality, their geographical scope and their relevance for transport (direct or indirect). This generated the classification of drivers and issues in the eight-fold scheme comprising demographic factors, attitudes, social policy related developments, institutional arrangements, science and technology, environment, economy and politics. It also provided much of the knowledge base for the specification of the transport impact pathways at the micro-level.
Scenarios represent visions/images of the future and courses of development organised in a systematic and consistent way. Participants to the expert panel consultations organised by the 'Foresight for Transport' project were provided with a summary of the issues produced through brainstorming in their respective sessions and asked to select those that would in their opinion be most suitable for describing (a) the present situation and (b) possible futures. Thereafter they had to individually and/or in groups elaborate textual descriptions of the baseline scenario and possible alternative futures. Through deliberation each group came eventually to agree on the baseline scenario and a limited set of alternative futures. The core research team of the 'Foresight for Transport' project then used the various scenarios elaborated by the thematic expert consultation panels to specify a generic baseline scenario and seven alternative future scenarios.
A Delphi survey involves a survey of expert opinion (structured in the form of a number of successive waves or rounds) in order to identify developments and/or trends and reaching gradually a convergence of opinion without physically coming together. The 'Foresight for Transport' project used a two-wave Delphi survey over a period of nine months, January to September 2003. The first-wave questionnaire was distributed among 455 experts around Europe with a specialisation in either transport and/or another field relevant to transport and mobility. This sought to validate the results of the first round of the expert consultations on critical non-transport factors and scenarios. Respondents were asked to rate critical factors with regard to their relevance; scenarios were assessed with regard to likelihood and desirability. The second wave questionnaire fed back to respondents the results of the first wave of the survey and asked them to reflect upon or confirm their original choices. Against this backdrop, several transport impact pathways were submitted for comments and validation. The second part of the questionnaire presented respondents with a series of quantifiable transport-relevant trends and asked for assessments with regard to the continuation (or reversal) of these trends in the short-term (2004-2010), medium-term (2011-2019) and long-term (2020+). The average time for completing the first wave questionnaire was estimated at between 45 and 60 minutes, that of completing the second wave questionnaire at between 90 and 120 minutes. The response rate for the first wave of the survey was 36 per cent, with a total of 164 responses. The longitudinal response rate was 65 per cent: 94 respondents reacted to the second wave of the questionnaire.
Trend extrapolation was used in 'Foresight for Transport' for indicators selected as being important for monitoring future developments with regard to either transport or external developments. The trend extrapolations were used in conjunction with the feedback provided by the Delphi survey to detail the reference scenario.
Running the exercise
The Figure below displays the management of the 'Foresight for Transport' study process as it leads towards the substantive outputs.
It is worth highlighting the following points:
- Scenarios are suitable primarily to enable a better understanding of the contextual conditions influencing policy or socio-economic developments. Policy recommendations must be embedded in this context but be based on assessments carried out at the micro-level.
- External or specialised expertise was used at various stages and through various technical means of the 'Foresight for Transport' study. Expert panels provided background analytical information necessary for understanding key parameters. Scenarios as well as impact pathways were validated by experts through the Delphi survey. Experts were also drawn into consultations for refining the impact pathways.
- The study relied primarily on qualitative methods. Contemporary trends were however described with reference to measurable benchmarking indicators. The expert assessments of how these trends will develop in the future (surveyed in the second wave of the Delphi) were analysed from both the qualitative and quantitative perspective.
The principal characteristic of the 'Foresight for Transport' process management was however less the combination of quantitative and qualitative methods but rather the inter-linkage between analytical steps – from description to validation and analysis and, especially, from the micro- to the macro-level and back – through the techniques in use by the project.
The reliance on external expertise, together with the requirement to cover as many alternative views as possible, means that a key element of a Foresight exercise, and its success, lies in the adequate and representative selection of experts and stakeholders to participate in the workshops and/or Delphi survey. Much of the effort that went into the 'Foresight for Transport' study has to do with trying to achieve a reasonable balance between broad coverage, on the one hand, and detailed knowledge, on the other.
Experts were selected on the basis of their work in the areas under investigation, as documented in academic and other publications, their membership of key policy communities (as decision-makers, advisers or members of lobbying organisations) and/or experience with working with business and industry. The structural constraints of organising a Foresight exercise (with regard especially to time availability for participating in workshops) meant that our sample was least representative of persons working directly in business or industry as well as officials working for European institutions. The indirect involvement of many of the other participants in these sectors, however, has controlled, in part, for this problem. In any case, this is an issue that needs special attention when planning and implementing a Foresight exercise.
The 'Foresight for Transport' Project produced:
- Six consultation documents – based on the brainstorming and scenario-writing carried out in the framework of expert consultations
- Two scientific reports on transport impact pathways based on the analytical work of the core team using the results of the expert consultations and Delphi survey as well as trend data.
- A monitoring system with metadata on the areas under investigation (see Figure above in the 'managing the process' section).
The 'Foresight for Transport' project mobilised a large number of experts in fields of relevance to transport policy through its strategic workshops and the Delphi survey. The experts came from academia, business & industry, civil society and policy administration. The project promoted networking among these actors as well as cross-sectoral integration.
Implementation and follow-up
The dissemination of the project results was primarily the task of the project team. The sponsors of the research (the European Commission) also disseminated the project results through CORDIS and its own Website on transport research. Employees of the European Commission participated on the expert panels. The final project results were presented to an internal seminar organised by the European Commission and targeting its own staff.
Hence, the results of the project are being disseminated through the project's own Website as well as other Websites, academic publications, publications in newsletters as well as conference/seminar presentations. Among others, the results of the project were extensively reported in a strategic overview report of the 'European Foundation on Working and Living Conditions'.
There has been no systematic follow-up of how the participants to the expert consultations/Delphi survey might be using the project's results or the foresight methods. However, reports from some of the participants point in this direction.
Our study focused on unveiling how changes within the external or policy environments come to impact on transport and mobility. The key term here is 'how', hence also the organisation of our research around the notion of 'impact pathway'. In other words, while models work with already established assumptions about cause and effect in order to achieve an estimation and valuation of impacts, our study has sought to clarify the cause-effect relationships as processes – in time and through a range of intermediate variables or policy domains.
It is this emphasis on the process rather than the quantifiable outcome that has led us to choose foresight as a methodology for the present study. Understanding process in conceptual terms, i.e. the impact pathways, implies integrating specialised knowledge as well as different normative appreciations regarding the future.
Thinking about the future does not only mean thinking about developments which are uncertain. It also means thinking about issues which are contested at this particular point in time. Foresight is a methodology that contested views to be reported and organised in such a way as to allow their further analysis, and so lead towards the identification of compromise solutions, which is what policy is ultimately about in democratic societies.
To summarise: the Foresight methodology is a good methodology when concentrating on issues which are process-oriented either substantively in terms of scope and content and/or with regard to the actors involved. Understanding how transport and mobility develops in the future and how these developments relate to developments in other policy domains is one such area. The knowledge thus gathered is of use for scrutinising and thereafter refining policy implementation strategies – especially at the micro-level – as well as for elaborating long-term strategies in a strategic manner, i.e. in relation to other policies. Furthermore the information gathered can be used to improve the assumptions underlying strategic models thus contributing to the amelioration of the latter's projections and their better interpretation.
What a foresight exercise cannot deliver – and what it should not be expected to deliver – is quantifiable and objective evidence-like statements or projections about short-term developments or impacts. It is in fact questionable whether there are any scientific methods that can deliver such straightforward and certain answers. Consider, for example, the question whether railway investment can be expected to lead to a modal shift from road to rail. We can probably answer this question with 'yes' but this alone is of little use in terms of defining a policy and implementation strategy. The more significant question is 'by how much' and 'will it be enough' [to effect a sustainable re-balancing of modes]. It is here that the answers begin to be more complicated and ultimately frustrating from the point of view of the policy-maker. Even if we were to assume that the quantifiable effect of railway investment on the modal split will be proportional to the amount of railway investment, there are several other factors that need to be taken into account for the estimation. These include at least the pricing regimes in place at present and at the time of completion of the infrastructure construction, as well as the level of development of the road network, including road-relevant technologies. In other words, there can be no single objective answer to this question, even if we were to assume that the policy question under investigation involves only a few parameters. At best one can provide an answer in terms of a 'range of values' that can in turn be used to better specify the degree of investment and the time framework of its implementation.
Keeping with the same hypothetical question the following can be said about the comparative advantage of foresight exercises as compared to policy assessments driven by modelling. Whereas models that work with a certain limited set of parameters can supply these ranges of outcome values, Foresight exercises can provide an insight into how the underlying processes work out in practice and, significantly, how these are influenced by other contextual conditions. The latter are especially of importance in situations where, say, the capital investment resources necessary for effecting a significant modal shift are simply not available. In such a situation the significant other factors, like, for instance, the logistic organisation of existing railway services or urban planning, might gain in significance in being more amenable to change at a lower cost or at a cost which can be easier distributed among public and private actors.
The problem of modern policy-making as well as science for policy is less that developments at the macro- or aggregate level have become more complex but more that the degree of conditionality at the micro-level has decreased together with the increase of autonomy of individual actors. Substantive and long-lasting transformation can thus only come about through the accumulation of several smaller-scale and often disparate – at first sight – actions operating at the interface of policy domains rather that within any single perspective. Foresight exercises, unlike mainstream transport modelling, can help identify such interfaces. Furthermore, in following the network logic they help bring relevant actors together, thus assisting in the drawing of interconnections at the level of action.
- European Foresight for Transport website (http://www.iccr-international.org/foresight/)