Strategies, missions and whole of government action
Every government is, in reality, a flotilla of many departments, agencies, tiers rather than a single thing. But all aspire to greater coherence. ‘Whole of government’ approaches - that mobilise and align many ministries and agencies around a common challenge - have a long history: during major wars, and around attempts to digitize societies, to cut energy use, to reduce poverty and to respond to the COVID-19 pandemic. These have been described using different terms - national plans, priorities, strategies and missions – but the issues are similar.
This paper, linked to a European Commission programme on ‘whole of government innovation’ looks at the lessons of history and options for the future. Its primary focus is on innovation, but the issues apply more widely. The paper outlines the tools governments can use to achieve cross-cutting goals, from strategic roles to matrix models, cross-cutting budgets, teams, targets and processes, to options for linking law, regulation and procurement. It looks at partnerships and other structures for organising collaboration with business, universities and civil society; and at the role of public engagement. It warns against simplistic, generic approaches (which are very common); over-reliance on committees (also very common, and rarely very effective); and too much focus on structures rather than processes, cultures and relationships.
1.Background
How should governments organise for cross-cutting goals, such as achieving net zero or coping with an ageing society? And how should they organise innovation of all kinds in ways that align with those goals and missions?[i]
There are hundreds of interesting examples to draw on, including many decades of practice in countries such as the Netherlands, Japan, France, Finland and Sweden, and recent innovations from Austria to China.[ii]
The terminologies change – from nation-building to strategy, problem-solving to missions[iii] - but the content is generally similar, covering how to stretch, guide and motivate all the many parts of government to move in a similar direction.
The easy default is to set up cross-cutting committees and boards of various kinds. But these are rarely enough to embed the goal, mission or strategy. More effective approaches usually combine attention to:
· Structures – agencies, departments, and roles
· Processes – for allocating resources, purchasing technology or adjusting regulations
· Relationships – within government and with key stakeholders
· Cultures – engaging and motivating the many people inside government and outside whose actions need to change if the broader objective is to be met.
Governments generally over-estimate the importance of structures and under-estimate the importance of processes, relationships and cultures (as do management consultancies and other advisers, partly because they are easier to summarise in neat powerpoint slides). In this paper I discuss how very different approaches are required depending on
· How much government itself is in control of the key actions: goals relating to business productivity require very different methods to those primarily involving public services (eg education) or behaviour change amongst the public, or outcomes such as poverty.
· The nature of the task (how novel or familiar, evidence-based or experimental)
· The political realities of government itself (how unified or divided, the relationships of different tiers)
· The organisation of finance (how long-term, cross-cutting, and how much ‘change margin’ etc)
2. History – burning platforms and political leadership
Many governments have attempted to achieve coherence, including in the ways they support innovation. The most dramatic examples come in times of war – for example, when countries including the UK, USA and Germany reorganised industrial production to meet urgent needs for tanks, aircraft and munitions. Ukraine is a contemporary example.
The US in WW2 is perhaps the most striking story of success. In the late 1930s, its military was run down. Yet the US went on to produce two-thirds of all Allied military equipment in WW2: 86,000 tanks, 2.5m trucks, half a million jeeps, 286,000 planes and 41 billion rounds of ammunition. This was achieved with continuous and rapid innovation alongside extraordinary mass production[iv], with radical breakthroughs in rocketry, radar and the jet engine, and the extraordinary scale of the Manhattan Project (which produced the atom bomb, pictured below) which at one point had an economic scale equivalent to the entire US auto industry.
Germany’s rocketry programme in the 1930s and 1940s was equally dramatic, described by one historian as ‘state mobilisation of massive engineering and scientific resources for the forced invention of a radical, new military technology’.[v]
The existential threat of war forced governments to ditch usual operating methods and provided a compelling common purpose, and a willingness to mobilise huge resources. Governments created new roles with authority to coordinate; reshaped budgets; and sped up decision-making, rather than relying on existing bureaucratic structures[vi]. There are also many peacetime equivalents, from the vast expansion of science and technology funding in the US prompted by the Cold War, to the strategy for digitisation in France (pushed by commissions like the Nora-Minc report in the 1980s and giving rise to initiatives such as Minitel).
Another striking example, highly relevant to the challenges of the 2020s, was the rapid acceleration of energy efficiency technologies in Japan in the 1970s in response to the sharp rise in the price of oil, and the sense that this was a profound crisis for a country that was so dependent on imports for its energy and the manufacturing industries on which its economy rested. This strategy achieved remarkable reductions in energy use and improvements in efficiency over a sustained period (though less was achieved in the 1990s and 2000s).
COVID-19 is the most striking recent example. The pandemic prompted a dramatic shift of global research activity (8.6% of all global research outputs in 2021)[vii], concerted action on vaccines and streamlining decision-making processes, and some innovative approaches (such as WirvsVirus in Germany and Hack the Crisis in Estonia).
In all of these cases, there was a sharp sense of threat which prompted political leaders to set a clear direction for government that was enforced across multiple departments, usually with the close engagement of business and involvement of senior figures from business. In each of these cases too attention was given both to innovation and adoption. [viii]
3. Tools for whole of government action
In the cases described above, innovation formed part of broader Whole of Government strategies - whether to reduce energy dependence or promote digitisation. Most governments are organised primarily in vertical silos, as they were in the 19th and 20th centuries. These may be organised by function (health, defense…); by population group (children, elderly); by profession (police, schools); by outcome (competitiveness, security); or by identity (religions, minorities). Within these structures, most governments also organise by method (statisticians, finance) and by place (region, locality, community).
However, there is nothing inevitable about this and many organisations in other fields have organised much flatter and more flexible structures, with more of a mix between verticals and horizontals, permanent functions and time-limited project teams.
Here I summarise some of the many tools which can be used to build up the horizontal capacity of government,[ix] and the insights from some of the guides that governments themselves have produced.[x]
In principle, the centre of a government can use any mix of command, incentives and persuasion to align the actions of multiple ministries and agencies. However, national governments vary greatly in their ability to use command and in their degree of centralisation – ranging from coalitions of parties with widely divergent views, to ones led by strong leaders with parliamentary majorities or autocrats. Usually, political pressures mean that departments have divergent aims, constituencies to please, and internal logics. Moreover, even where there is clear policy direction, there are likely to be very different levels of capability in different parts of government. [xi]
Coordination is also not always a good thing. It generally follows the shape of an inverted ‘U’: no coordination leads to poor outcomes, but so does too much coordination, since that invariably means too much time spent in meetings or communication with others inside government rather than focusing on action.
Within governments, too, there is some advantage in having duplication, competition and incoherence. Harold Wilensky famously wrote in his book ‘Organisational Intelligence’ about how FD Roosevelt deliberated created ‘constructive rivalry’ within his administration. Ministers often had overlapping roles, or were given similar tasks, so that any claims would be checked and challenged by others. Roosevelt ‘structured their work so that clashes would be certain’ which had the effect of ‘the timely advertisement of arguments, with both the experts and the President pressured to consider the main choices as they came boiling up from below’. This, Wilensky argues, was much better than the groupthink, conformity and sycophancy that often characterises the centres of governments.
The key is to achieve a balance. Governments only organised around top-down missions or strategies will be as ineffective as ones divided into multiple competing ministries.
Coordinating committees
The default option for any government is to create a cross-cutting committee or board, and these can have some impact: much of the work of government is helped by committees of all kinds. Many countries – from Japan and Denmark to the UK – have, for example, created committees chaired by the Prime Minister to coordinate and drive science and technology policy, or to guide government priorities.
However, experience suggests that more systematic institutionalisation of coordination is usually needed. Committees on their own rarely have much impact on the underlying powers and currencies of government – budgets, jobs, laws and departmental cultures – and their creation is sometimes used as an alternative to action rather than as a means of ensuring action. At best they are a necessary rather than sufficient condition for change.
Central teams
Many governments have teams at the centre charged with coordination on important issues. The White House Office of Science and Technology Policy (OSTP) is a good example, with experienced and expert staff, though often only limited influence over budgets in departments. The Vaia team in Slovakia is another good example, that’s acted as a creative catalyst across government.
Organising around strategic priorities
An option that goes well beyond committees explicitly creates a small number of strategic priorities and concentrates power and budgets around these, pushing the more functional departments or agencies a level lower in the structure. This is common in cities which have a number of vice or deputy mayors, and it was the approach adopted by the European Commission in the middle of the last decade, with the creation of Vice Presidents. A similar approach has been taken in some countries with the appointment of super-ministers, though the option of creating a strategic level above departments generally works better as the super-ministers become pulled into too much operational detail. These can make it easier to align R&D with policy across several domains.
Some of the historical examples essentially used this model, authorising new roles to provide strategic oversight, such as the US’ appointment of William Knudsen to oversee military production in the 1940s, or more recently Portugal’s appointment of Admiral Henrique Gouveia e Melo to coordinate its COVID-19 response and vaccination programme.
Matrix models and system connectors
Some governments have experimented with partial matrix models. So, alongside the vertical silos some functions are organised in cross-cutting ways, giving the individuals in these roles a dual accountability, both vertically upwards and horizontally outwards. This can work for finance – with finance teams with a department answerable to their heads of department but also connected to the finance ministry; it can work for professions (such as statisticians, economists and planners); and it can work for digital government (and has been used in many governments as a way to link together the teams working on issues such as payments, identification or government services).
Another approach, adopted by New Zealand, creates roles within the bureaucracy whose job it is to weave the threads together, in this case with the title ‘system leads’ to signal their explicit responsibility to take a whole of government approach. These roles are essentially matrix models, with the roles held by senior officials who also have vertical responsibilities. The combination of vertical and horizontal responsibilities in a fairly small government makes it quite easy to negotiate.
Digital platforms and shared modules
The importance of digital technologies to the everyday work of government has prompted many countries to adopt whole of government approaches. Over the last few decades the norm is often for agencies and departments to commission their own software, websites and services. This remains the case in many countries but leads to obvious problems. One is interoperability – services unable to communicate with each other (eg emergency services). A second is waste: with often hundreds of separate websites, and sometimes dozens of separate systems for payments or authentication. For two decades there have been moves to streamline and standardise: in the 2000s with the rise of personal accounts and integrated payments systems in some countries; the principle of ‘tell me once’, so that once a citizen had shared information with one part of government it would be available to others.
Service New South Wales in Australia is a particularly good example of creating a horizontal function that connects multiple agencies, and so makes life easier for citizens.
In the 2010s, some countries went a step further with the rise of systems like Estonia’s X-road (now in use in over 30 countries) which provided a shared platform both for government services and for the private sector with common systems for authentication. Another approach was the creation of integrated Government Digital Service teams, with a remit to create modular and shared methods for tasks such as payments. The largest example of all has been India’s Aadhaar (under its Unique Identification Authority) and ‘India Stack’ which provided a common ‘digital public infrastructure’ for payments and authentication across a wide range of services. Such whole of government approaches may also be relevant to the use of AI – with at a minimum some sharing of expertise, and common approaches to procurement.
Other methods
Other options redirect the many everyday tools of government organisation to better align with a cross-cutting goal. These can be powerful ways to shift behaviour. They include:
· Cross-cutting structures, so that alongside classic functional ministries new central units are created with a cross-cutting remit (such as the various government digital units or the unit Ireland recently created on child poverty, drawing on the experience of Scotland and New Zealand)
· Cross-cutting roles for ministers and officials, again combining classic vertical roles with ones that have a cross-cutting dimension (for example, focused on competitiveness, social exclusion or AI)
· Cross-cutting or pooled budgets that focus on an issue or outcome rather than being allocated to ministries in the first instance
· Shared targets that are spread across multiple ministries, a common approach for governments that use quantitative targets as a planning tool
· Shared data (for example connecting health and care) and shared knowledge (for example a pooled research function, rather as JRC does for the European Commission)
· Cross-cutting foresight and strategy teams with a remit to map out both potential and desirable futures on issues such as energy, obesity or AI
· In-house consultancies – for example around data, policy design or digital (such as the various Government Digital Services which became common in the 2010s, or the new centralised consultancy created in the Australian government in 2023). These can act as glues across the system – though their effectively depends on their skill in handling relationships and conflicts, and the inevitable pushbacks.
· Cross-cutting implementation teams – for example, on topics like digitising public services or cutting homelessness
· Double keys – which give a central official or minister effective veto powers over budgets or implementations so as to encourage strategic alignment (for example to ensure interoperability of software: this approach was used for digital in the UK government in the 2000s)
· Local partnerships - which are very common in urban regeneration (Bilbao Metropoli 30 is a particularly famous example), typically linking several tiers of government, civil society and business, and can include formal shared quantitative targets
· Local cross-cutting outcome budgets - these attempt to integrate budgets at a local level to enable preventive action which in time will save money for government. Justice Reinvestment is a fairly widespread example, with schemes to fund programmes to divert crime, repaid by savings in courts and prisons.
· Cross-cutting case management, one-stop shops and multi-disciplinary teams and other tools to ensure that multiple professions and agencies work in a coherent way to support clients or places
This is a very brief summary of a wide range of practical methods that are used to bolster collaboration.[xii] They provide a first menu of options for any government wanting to get beyond reliance on committees. The diagram below summarises them.
Individual qualities matter greatly. Some of the boldest initiatives depended on exceptional leaders who could motivate and drive change, often against bureaucratic obstacles.
Perhaps the biggest challenge for anyone designing new structures is how to avoid the risks ‘organogram illusion’ – the common belief that a neat organogram on powerpoint slides represents reality. Instead, in practice, just as much depends on the calibre of the individuals appointed to key roles, their relationships and purposes.
But the other challenge is that coordination and cooperation is costless. I often argue that we should think of this as an inverted U (summarised below): no cooperation tends to lead to poor performance. But so does too much, when people spend all their time in conversation with others. The key is to find the optimum, and to reduce the friction and cost involved.
4. A taxonomy of whole of government approaches
We can summarise some of the key design choices facing any government seeking to implement a more strategic, whole of government approach, and how to decide what approach will work best for which tasks. The key issues concern a series of resources: power, knowledge, authority money and trust.
Power: how much government control? A basic question is whether government has much direct power over the phenomenon being influenced. In the case of a war, government does at least in theory have direct control over its armed forces, and often over supply chains, or digital procurement. But where government is seeking to influence the economy, it depends on much more collaborative partnerships that share power both to design the strategy and implement it. This is where examples like the industrial policy approaches in Germany, Japan, Taiwan and south Korea, or the US strategy around semiconductors, are relevant. In other cases where public services are critical, government is likely to have more formal powers (though again much will depend on its ability to persuade others to align their actions).
Knowledge: how much evidence or certainty? A next question is whether there is reliable knowledge about what to do. If there is strong and reliable evidence on how to, for example, improve public health, literacy or poverty then a strategy can be designed, supported by law and budget allocations. But where the means are more uncertain (as with very fast changing AI options) then a strategy needs to include experiment and discovery.
Authority: how much political capital? The practical options for whole of government depend greatly on political realities. Does the political leadership have a strong position – a large majority, a coalition with broadly convergent views? Or is it precarious or divided? What too are the relationships between different tiers of government?
Money: how much finance? The options for governments are very different where there is free money to devote to a cross-cutting goal: a ‘change margin’ of resources not locked into existing structures or commitments.
Trust: how much willingness to collaborate? Whole of government action is driven by trust. The benefits of cooperation and collaboration of all kinds have an inverted U shape. Too little leads to bad outcomes, but so does too much. So, a key question is clarifying the minimum efficient number of players needed to be aligned. For example, if the aim is to rapidly increase housing provision, then there will need to be alignment of planning, housing, big municipalities as well as major developers. At some point it will also be vital to involve education (to provide new schools), health (to provide new services) and so on: but their involvement will not be essential early on and can slow down action.
5. Whole of government innovation
In this section I look further at how governments have used ‘whole of government’ approaches for innovation. I’ve already mentioned some striking examples, including the US in WW2, Japan in the 1970s and France in the 1980s. Traditionally policy for science and technology has been guided by four main logics that combine purposes with forms[xiii]:
State interest: the longest tradition is that of mobilising science and technology to serve the interests of the state, primarily for the military (military R&D continues to account for roughly half of public funding in the US), but also for surveillance and sometimes for the subordination of society. State interest doesn’t always align with a broader public interest, though sometimes, particularly in democracies, it does. Such military missions remain important and have become more so in the 2020s as geopolitical tensions have risen.
Economic/industrial: during the late 19th century, and the 20th century, states increasingly used science and technology to fuel economic growth, with funding aligned to economic and business interests, and close engagement of major companies. A recent OECD survey helpfully analysed the many forms industrial strategy has taken,[xiv] from attempts to direct technology and industrial structure to much more open approaches. This continues to be the dominant lens through which new scientific and technological breakthroughs – from life sciences to quantum – are seen.
Autonomy: the third, very different approach, puts control over funding, and the setting of both ends and means, in the control of scientists themselves, guided by peer review. This became a dominant approach after WW2, partly in reaction to the excessive control of Stalinism and Nazism, and partly the result of belief that autonomy for scientists would in the end deliver better results for society.
Social goals: the fourth option of prioritising social and broader political goals has been much rarer, with the important exception of healthcare. Relatively few governments have had developed systems for fuelling innovation in relation to topics such as education, welfare or mental health. Although there are pockets of work on ‘social R&D’ (for example in Canada and Australia) this remains unusual in Europe. This may help to explain why analysis shows major misalignment between SDG priorities and STI funding, as for example in this recent analysis from the STRINGs project (pictured below).[xv] It also explains why the social, environmental and behavioural aspects of health have generally received much less attention and public financial support than pharmaceuticals.
However, recent shifts in policy have better integrated economic, environmental and social goals. The Netherlands Top Sector approach is a good example, linking several funding streams, the Regional SME Innovation Incentive scheme (MIT), Applied Research Organizations (such as TNO) and the Dutch Research Council (NWO-PPS), with an emphasis on following a private sector lead through topping up private investment.
This approach, which has been highly successful in driving innovation on cross-cutting goals, was given extra energy in 2019 by the 20bn Euro National Growth Fund providing additional public funding for projects aimed at long term economic growth and, increasingly, positive impact on societal challenges and transitions. This connects the Ministry of Economic Affairs with other departments (e.g. Infrastructure, Health, Defence), and with an emphasis on the Netherlands position in global supply chains.
This example is unusual in part because of the way it combines breadth of aims with a sharp focus on the most promising ideas. It’s also unusual in linking a focus on hardware with a broader approach to business models and services (echoing an excellent recent piece by the OECD which warned of the ‘STI-only trap’ in net zero missions: an excessive focus on physical technologies which risked undermining the success of such missions).[xvi]
6. Potential priorities for Whole of Government Innovation
What might be the priorities for Whole of Government Innovation in Europe over the next decade? And what different methods might be relevant? As indicated above there is a long history of governments seeking to prioritise particular ends such as digitisation or energy efficiency, and in recent years the language of missions – actively promoted by mainly US commentators in the 1990s - has again become popular, prompting greater attention to how they can be achieved[xvii] and providing a new label for cross-cutting goals. A separate body of work has been developing on how to manage transitions.[xviii]
For many governments, a first priority for ‘Whole of Government Innovation’ is to better orchestrate support for science and technology as a whole, ensuring that policies from one department (eg those concerned with immigration, tax or university finance) do not discourage research and development. This can be helped by having central units with a remit to reduce the risk of contradictory or clashing policies, and to keep track of data on issues such as brain-drains or commercialisation or the emergence of innovation districts.
Beyond this broad objective, the likely priorities for governments over the next decade which can be roughly distinguished according to whether they are:
· Focused on desired outcomes (like net zero). Decarbonisation strategies can link research on new energy sources, storage, transport housing, all the way through to regulation, taxation, trade. Because this concerns the whole of a carbon-based economy it has to be holistic.
· Focused on broader systems change goals, such as transforming food systems to reduce dependence on meat and using the full gamut of tools from research funding to regulation and provision of school food.[xix]
· Focused on inputs such as artificial intelligence, encompassing governments’ own role in commissioning AI (for schools, social security, health, criminal justice and tax), as well as institutions for regulation, upstream R&D and more.
· Focused on key fields of technology such as quantum, with relatively little attention to potential uses.
· Focused on specific strategic priorities such as more autonomy in semi-conductors (a major priority in the last two years for the US and the EU).
· Focused on distributional goals such as greater equality between regions or social goals such as reducing homelessness or anxiety.
These may overlap. An AI strategy may be simultaneously about research, positioning, commissioning for public services and about adoption.[xx]
7. Public engagement and democracy
Science and technology policy has generally not had much direct public engagement. Instead, the priorities have been set by governments, by business or by scientists themselves. However, there is some evidence that this is creating problems – eg in the UK a majority say that public R&D doesn’t benefit people like them; only 9% believe it benefits them a lot.
There have been many past attempts in some countries to mobilise public input on difficult ethical issues – notably in Denmark – with citizens juries, ‘consensus conferences’ and deliberative processes around issues such as fertilisation and embryology.
In some countries the work around roadmaps and transitions has also involved a much wider group of stakeholders, including small business and civil society.
The Netherlands is an interesting example of a recent attempt to mobilise public input on broader priorities. In 2015 the Dutch Minister for Education and Science and the Minister for Economic Affairs launched a task force to shape the science agenda, involving research institutions and business. They decided to engage the public, asking them to pose questions they wanted to see addressed by science, with the proviso that these had to be solvable within ten years as well as challenging and ground-breaking.
The process was widely communicated including through the countries’ most popular daily television talk show, music festivals and TED talks. 11,700 questions were submitted, which were then clustered into 140 key questions and 25 routes for science, such as ‘sustainable production of safe and healthy foods’ or ‘Resilient and relevant societies’. [xxi] Since then, these have guided the National Research Council, which funds inter-disciplinary research to address ‘wicked-problems’ in society as well as other thematic programmes.
8. Implementation with learning
Whole of government programmes need serious attention to implementation even more than programmes within departments. Within departments there is often plenty of experience with the practicalities of ‘how’ as well as ‘what’: how money actually achieves impacts; the skills and capabilities needed for new tasks; the potential resistance. For more complex, cross-cutting goals this will be harder. This is why it’s important to build in processes for rapid learning and experiment, for example setting short-term goals and using 100-day sprints for rapid innovation and problem solving; ensuring it’s clear whose responsibility it is to gather data and ensure learning from unexpected patterns and results; investing seriously in capability; and consciously bringing in the requisite skills from any source.
There are many areas where this kind of implementation preparation is missing, including much of cross-government digital work, from customer interfaces to procurement of AI; and it has often been missing in infrastructure planning.
However, some of the work on pathways and roadmaps in individual countries (on decarbonisation, the circular economy and other topics) addresses these issues head-on, often involving close collaboration between governments, businesses and sector organisations, and collaboration between multiple tiers of government with integrated teams to plan, implement and learn. But, as indicated before, these aspects of implementation tend to be underestimated in governments which assume that with the right structures, or policy statements, the challenges of implementation will be solved.
9. Moon-shots as a tool for innovation policy
A very different approach has emphasised top-down goals, with some governments using the language of moon-shots, echoing US strategy with NASA in the 1960s, when a very bold commitment, and extraordinary financial resources, enabled the US to land a man on the moon before the end of the decade.
The methods used for moon-shots are most relevant where there is a very specific objective and a specific potential technology solution, as with the original NASA Apollo missions. However, in most of the cases listed above as potential fields for ‘Whole of Government Innovation’ this parallel is misleading, since the objectives are much broader (such as CO2 reductions in a particular industry), require a wide range of complementary solutions rather than a single technology, and are likely to have implications, some threatening, for many more existing interests.
In this respect strategies around cancer are as relevant as the Apollo missions in terms of offering insights. The US ‘war on cancer’ was announced in 1971 and began by seeking a single magic bullet. Nixon promised that “the same kind of concentrated effort that split the atom and took man to the moon should be turned toward conquering this dread disease” and hoped that this would be achieved by the time of the American bicentennial in 1976.
However, a review published in 1986, concluded that “some 35 years of intense effort focused largely on improving treatment must be judged a qualified failure”[xxii] while an updated version by the same authors a decade later in the late 1990s was titled “Cancer Undefeated”.[xxiii]
Old habits die hard, however, and in 2016, Joseph Biden, then Vice President, launched a “Cancer Moon-shot” initiative, which once again promised to end cancer. Many in the field felt ambivalent. Half a century had brought extraordinary advances in understanding of the complexities of cancer, its genetic aspects, and the roles of environment, but they had also made the political rhetoric of moon-shots look simplistic and misplaced.
That said, the language of moon-shots can be useful for communications and motivation, even if it is less useful as a guide to action. A good example is Japan’s current Moon-shot R&D programme (see below) which uses the word as a broad metaphor for ambitious, stretching goals.[xxiv]
10. Risk, impact and portfolio approaches
The alternative to very focused moon-shot approaches is to aim at a portfolio of initiatives and projects addressing different aspects of a problem (and some current initiatives that use the moon-shot language are in fact much more like this). Such portfolios are likely to include:
· Upstream research – ensuring that research funders are supporting basic research aligned with the broader objectives
· Downstream implementation – ensuring coordination in relation to technologies coming into markets, or into public procurement
· Policy – aligning policies, from subsidies and tax to law, both to remove obstacles and sometimes to enable adoption
· Regulation – aligning regulatory rules to enable adoption of key technologies, and to mitigate risk
· Procurement or commissioning in fields where government plays a major role – such as health and education
· Advocacy, education and socialisation of new ways of thinking and acting
As summarised in the diagram below, in relation to research a portfolio may include some projects with high risk and also high potential impact, alongside others with lower risk but lower likely impact. The portfolio may cover very different pathways and timescales.
The job of managing projects and programmes involves seeking to increase their impact and reduce their risk over time, as ideas are developed and refined and as more data is collected on impacts. In other words, the aim is to move projects downwards and towards the right over time (and, at least in principle, a portfolio can be regularly reviewed through this lens, for example using technological readiness measures):
The variety of tools that governments can use for experimentation are also relevant here in part because these are ways of building up more evidence about effectiveness as well as growing capacity.[xxv] A group of experiments on a topic such as incentives for home retrofitting, or large-scale mental health provision, should generate insights into how these are best designed and their potential impact for different groups.
A key debate in the field is whether key societal missions can be thought of in linear, deductive ways, with over-arching missions then divided into sub-units, in a logical hierarchy. This approach is often appealing to governments and is easy to communicate. An alternative view, however, sees systemic change as inherently more messy and complex, better understood as a network of loosely connected actions (a recent piece by Wanzenböck and others provides a good overview of these different mindsets and methods).[xxvi]
11. The role of research and innovation funding agencies
All EU countries have some institutions responsible for funding research, often working alongside others responsible for technology, tech transfer, start-ups and individual sectors. Some countries have tried to create new agencies to take bigger risks, such as Germany’s Bundesagentur für Sprunginnovation, focused on disruptive innovation.
This ecosystem can be quite complex – with some parts operating with autonomy and others much more influenced by political direction. It is relatively easy for funders to prioritise a new field – such as energy efficiency or AI – but usually much harder for them to take an end-to-end approach, or to align their funding choices with regulatory decisions.
There are then a range of tools for funding[xxvii] which roughly break down between:
· Funding teams or units based on past track record and giving them freedom to decide how they work and what they focus on
· Funding projects, often with requirements for multiple collaborators, and more specification of both ends and means
· Funding whole institutions – particularly universities, or government research labs
· Funding outcomes, through challenge prizes and other open innovation tools
· Funding investment and enterprise creation and growth – through grants, loans, equity and convertible forms, with an emphasis on commercialisation
· Funding experimental spaces – labs, testbeds, accelerators
· Funding connectors – intermediaries of all kinds with a remit to spread knowledge
Some agencies such as Vinnova, and the Climate KIC, have experimented with new ways of supporting innovation in ways that form part of broader transition strategies. As indicated earlier, the Netherlands has also been a leader in engaging business in long-term strategic collaborations, for example, as mentioned above with their ‘Top Sector’ approach launched in 2011 to link business, universities and government. Singapore is also an interesting model with its ‘Alliances for Action’, industry-led alliances working with government to prototype new ideas.
In some countries, agencies that are one step removed from the core government structures can play a role of coordination. In Australia, CSIRO is an arm of government that has experimented with organising its work around missions. In Europe, funding bodies such as SITRA and Tekes in Finland have led on key issues, perhaps taking advantage of their relative distance from the everyday work of the big ministries. Sweden’s Vinnova’s ‘Strategic Innovation Programmes’ (see below) are an interesting example.[xxviii]
These agencies take care not to fall into the trap of excessive faith in performance management which is often found in the English-speaking world, and is popular amongst consultancies (the belief that if you only set clear targets, trajectories and reporting methods, you can achieve complex objectives). Instead, they recognise that they are working in more complex ecosystems where there is rarely a simple causal relation between actions and results.
12. Net zero and Whole of Government coordination
Many governments around the world have tried to create cross-cutting structures to help with strategies for net zero. These involve some alignment of law, structures, financing models and reporting systems, though it is too soon to judge which have been most effective, since most of the successes in carbon reduction over the last twenty years predate these new approaches. A recent survey concluded that ‘most net-zero mission achievements have been realised in the definition of strategic agendas and in setting up governance structures. Evidence of joined up implementation remain rarer and small scale.’[xxix]
First, some of the laws. These include Sweden’s Climate Act 2017 which requires government to report progress to the Riksdag (parliament) every four years, Germany’s Klimatschutzgesetz (climate protection law) passed in 2019, France’s Loi Energie et Climat in the same year, and New Zealand’s Climate Change Response (Zero Carbon) Amendment Act 2019.
Next, the structures. Many governments have some kind of board or cross-cutting function, such as Ireland’s Climate Action Delivery Board within the office of the Taoiseach (the prime minister) composed of permanent secretary-level officials to oversee the delivery of the detailed Climate Action Plan, which lists more than 180 actions and specific measures.
Sometimes a lead role is played by an agency, as in Norway where the ‘climate cure plan’ has been co-ordinated by the environment agency since the late 2000s, drawing on inputs from many agencies. Next comes funding. Governments have used ‘satellite accounts’ for many years in relation to the environment, but none have fully integrated net zero into budget allocations, though there are attempts, such as in Finland, where the Finance Ministry has to report on progress towards net zero.[xxx] Other types of funding body can also play critical roles. The EIB is key now to providing capital across Europe; at a national level, funders like Germany’s state investment bank KfW can take the lead on funding of solar, other renewables and energy efficiency programmes.
Finally, reporting. Many governments now have comprehensive systems for reporting progress against targets. Some have taken this kind of embedding further. Since 2019 New Zealand has required all policy proposals for Cabinet to have a climate impact policy assessment.
However, few governments have yet fully integrated their science and technology systems. For several decades many have committed to broad missions around energy efficiency or the circular economy, and ambitious commitments to investment in green technology have been a staple of political summits since the 1980s. These have played their part in the sharp declines in the costs of renewables, but it’s often been easy to rebadge existing programmes as green or clean (what’s sometimes called ‘mission washing’), and these types of funding mission tend to be organised without much connection to the design of laws and regulations, or other policy instruments.
Moreover, much more progress is needed in the orchestration of knowledge about net zero – what works, when, why and how. Although there has been an explosion of repositories, observatories and scans of technologies and policies these remain quite hard to use.[xxxi]
In relation to structures, the UK is an interesting example of both progress and backward steps – in the late 2000s it was one of the first countries to create a comprehensive machinery in government including the 2008 Climate Change Act, with legally binding targets, a Department of Energy and Climate Change, and a Climate Change Committee to monitor progress. Later it also developed plans for a Green Investment Bank. However, all of these (except the CCC) were later dismantled.
13. Shared data and knowledge to support coordination
A precondition for effective coordination and alignment is shared knowledge. Without some shared picture of the current situation, the key tasks and trends, and the available knowledge of what works, it’s hard to embed coordination.
Over the last decade there has been a proliferation of new institutions to better share knowledge. These include the work of JRC in Brussels; the various ‘what works’ centres around the world (for example, focused on healthcare, education or employment); and repositories of all kinds, and there is a growing literature on how to organise knowledge infrastructures.[xxxii]
The IPCC provides a shared diagnostic picture of climate change patterns and bodies like C40 for cities try to orchestrate knowledge about interventions. In a recent paper I surveyed the many global sources of knowledge relevant to decisions about net zero.[xxxiii]
Within a government a basic requirement is some transparency on current actions – what money is spent on what purposes, with which institutions and in what geographies. Without such a comprehensive map it’s hard to know where there may be gaps, duplications or misalignments.
The diagram below, developed with Demos Helsinki[xxxiv], summarises an approach in which top-down actions are aligned with bottom-up initiative and innovation, and held together with common stores of knowledge, shared data and shared pictures of future possibilities:
14. Hearts, minds and networks
Governments can use a mixture of command, incentives and persuasion to achieve better coordination. People in governments, as elsewhere, favour coordination in principle, but are often not so keen about being coordinated by others. Governments can force directions on their civil servants and other stakeholders, but they are more likely to succeed if there is a sense of shared mission, diagnosis and commitment.
This is why the quality of individual leaders matters – the best ones can inspire, motivate and energise. It’s also why top-down initiatives need to be complemented by exercises to achieve buy-in at multiple levels of the government: large scale gatherings, discussions on cross-cutting goals and co-design of strategies may be essential to achieving sustainable, as opposed to cosmetic, engagement.
There is a huge literature on how this can be done, drawing on social psychology, knowledge management, organisational development and the experience of communities of practice, and examples such as collaboratives in the health field which engage large numbers at multiple levels to encourage a shared sense of direction and priorities.
Psychology can also provide insights into how complex, long-term tasks are best organised in large groups – ideally with ritual and theatre as well as rational analysis to make it easier for the participants to link actions taken now with the results achieved much later.
However, few of these methods have been used in relation to innovation in part because of the traditional separation between upstream and downstream funding, and the separation between research support and other government tools. But such methods may be increasingly important for many of the challenges of the 2020s.
The European Commission has begun to experiment with some similar methods – with the European Research and Innovation Days which began in 2019, and net zero may be an ideal topic to orchestrate the many communities playing roles, from scientists to policy-makers, business to municipalities, to achieve shared understanding and commitment.
15. Some conclusions
Coordinating the whole of any government is always bound to be difficult, unless there is a truly existential threat (as there briefly was during COVID). The complexity of approaches may simply be too large.[xxxv]
More often the realistic priority is to create flotillas or constellations which connect a small number of ministries, regulators, the finance ministry and the Prime Minister’s office, for example focused around a shared approach to energy, transport or public health. It may be preferable to achieve real engagement from a small cluster rather than rhetorical commitment from a broader group.
Indeed, there will often be a trade-off between focus and breadth. It’s interesting that, apart from in wars, the Mission approach has sometimes been most successful with dedicated agencies and little attempt to coordinate or align with other departments. In the US, examples including NASA and DARPA, were given freedom to operate outside conventional structures and with little direct engagement with potentially relevant ministries (DARPA avoided even aligning much with other DOD innovation spending). Greater focus may allow for more risk-taking and boldness. But it also makes these examples less relevant to the complex transitions many governments want to support in the 2020s.
In most cases too the key challenges of coordination are bound to involve actors outside government, including major businesses, universities, civil society and municipalities. This will be true of tasks like net zero and the circular economy and lies behind approaches like the Netherlands ‘top sector’ model, and Japan’s longstanding approaches to industrial policy and technology missions. Indeed, these are reasons why it may be more useful to think in terms of ‘whole of governance’, or ‘whole of society’ rather than just ‘whole of government’.
The temptation for governments and advisers is to seek a standardised approach, for example with cascading hierarchies of coordinating committees. However, it is more likely that effective arrangements will be tailored to task and context, to where the greatest pools of relevant knowledge are to be found, as well as the key concentrations of power and money. Overly generic approaches may be liable to fail. With that caveat, most governments would benefit from action in at least the following fronts:
· Creating some central capacity for leadership, and coordination – ideally with a direct mandate from a Prime Minister or President and bringing in significant outside expertise to provide visible leadership for programmes. Without this it’s unlikely that political capital can be translated into effective action.
· Curating key clusters or groupings, tailored to the task, ideally with clarity about goals, accountability and contributions
· Ensuring long-term stable funding for key agencies, preferably with rolling 5-10 year rather than annual budgets
· Supporting action focused partnerships with businesses and others, with their membership shaped by the task at hand
· Using cross-cutting digital approaches, whether to authentication and payments or procurement to avoid duplication
· Taking implementation and learning seriously – using sprints, cross-cutting implementation teams and clarity on targets
· Combining strategic clarity with space for radical innovation and experiment – challenges, accelerators, labs of all kinds
· Ensuring mutual transparency – with shared data, and knowledge about actions, funding, intended goals to enable coordination, and clarity about whose responsibility it is to curate shared knowledge and data.
· Bringing the people working on tasks together – and recognising that informal relationships of trust may be vital to getting the formal structures to work.
Notes
[i] https://ec.europa.eu/research-and-innovation/en/statistics/policy-support-facility/psf-mutual-learning-exercise-whole-government-approach-research-and-innovation.
[ii] Some recent literature can be found in these: Breznitz, D., Ornston, D. & Samford, S. (2018). Mission critical: the ends, means, and design of innovation agencies. Industrial and Corporate Change, 27(5), 883–896. https://doi.org/10.1093/icc/dty027; Janssen, M. (2020). Post-commencement analysis of the Dutch ‘Mission-oriented Topsector and Innovation Policy’ strategy. University of Utrecht, Mission-Oriented Innovation Policy Observatory (MIPO), https://www.uu.nl/sites/default/files/Post-commencement%20analysis%20of%20the%20Dutch%20Mission-oriented%20Topsector%20and%20Innovation%20Policy.pdf; Bronchi, I. et al. (2022). Missions for governance: Unleashing missions beyond policy. Demos Helsinki, https://demoshelsinki.fi/wp-content/uploads/2022/10/Missions-for-Governance-Demos-Helsinki.pdf The Mission-Oriented Innovation Network also provides a useful body of case studies and approaches: https://www.ucl.ac.uk/bartlett/public-purpose/partnerships/mission-oriented-innovation-network-moin
[iii] Prompted by books such as D. Osborne and T. Gaebler, Reinventing Government 1992. The mission language was very widespread in the 1990s, fell out of fashion in the 2000s and then was revived in the last decade, see eg European Commission, Directorate-General for Research and Innovation, Mazzucato, M., Mission-oriented research & innovation in the European Union : a problem-solving approach to fuel innovation-led growth, Publications Office, 2018,
[iv] From Arthur Herman, Freedom’s Forge, Random House, 2012
[v] Michael Neufeld ‘The guided missile and the Third Reich:Peenemunde and the coming of the ballistic missile era, New York, Free Press 1995 p51
[vi] A dilemma for any programme of this kind is how to achieve the right balance between continuity and change – existing structures, teams and processes will often be more efficient than new ones. So too much change and novelty can be as problematic as relying on structures that were designed for very different tasks. The challenge of getting the balance right is a common theme in a century of literature on leadership, management and change.
[vii] https://journals.sagepub.com/doi/10.1177/26339137221146482
[viii] See: Dan Breznitz, Darius Ornston, The politics of partial success: fostering innovation in innovation policy in an era of heightened public scrutiny, Socio-Economic Review, Volume 16, Issue 4, October 2018, Pages 721–741, https://doi.org/10.1093/ser/mww018
[ix] Mulgan, G (2008) The Art of Public Strategy, Oxford University Press; Jacob, K. & Volkery, A. (2004). Institutions and Instruments for Government Self-Regulation: Environmental Policy Integration in a Cross-Country Perspective. Journal of Comparative Policy Analysis: Research and Practice, 6:3, 291-309. DOI: 10.1080/1387698042000305211; Lægreid, P., Verhoest, K. & Jann, W. (2008). The Govenance, Autonomy and Coordination of Public Sector Organizations. Public Organization Review, 8, 93–96. https://doi.org/10.1007/s11115-008-0056-5; Lægreid, P. & Rykkja, L.H. (2022). Accountability and inter-organizational collaboration within the state. Public Management Review, 24:5, 683-703, DOI: 10.1080/14719037.2021.1963822; Lopes, A. V., & Farias, J. S. (2022). How can governance support collaborative innovation in the public sector? A systematic review of the literature. International Review of Administrative Sciences, 88(1), 114–130. https://doi.org/10.1177/0020852319893444
[x] https://www.publicservice.govt.nz/guidance/guidance-system-design-toolkit
[xi] This is why I emphasise ‘steering through capability’ as a key aspect of governance: https://demoshelsinki.fi/wp-content/uploads/2020/12/dh_steering-through-capability2020.pdf. Similar approaches became common in business in the 2000s, prompted by David Teece, Gary Pisano and Amy Shuen’s 1997 paper Dynamic Capabilities and Strategic Management, which focused on "the firm’s ability to integrate, build, and reconfigure internal and external competences to address rapidly changing environments." This drew in turn on Nelson and Winter’s 1982 book An Evolutionary Theory of Economic Change. More recently these ideas have been taken up in the public sector. See also Rainer Kattel: https://www.ucl.ac.uk/bartlett/public-purpose/publications/2022/mar/dynamic-capabilities-public-sector-towards-new-synthesis
[xii] An extensive overview of holistic governance approaches is: 6. Perri, D. Leat, K. Seltzer and G. Stoker (2002), Palgrave, Basingstoke; and Christensen, T. and Per Laegreid, The Whole-of-Government Approach, Regulation, Performance and Public-Sector Reform, Working Paper 6-2006, Stein Rokkan
Centre for Social Studies.
[xiii] My book ‘When Science Meets Power’, Polity 2024, sets out this history in much more detail
[xiv] https://www.oecd-ilibrary.org/docserver/0002217c-en.pdf?expires=1672739944&id=id&accname=guest&checksum=9D33E8D3DD855D5E1F5ABCAE68CAC072
[xv] STRINGs report: Steering Research and Innovation for the Global Goals, 2022
[xvi] https://www.oecd.org/greengrowth/2022GGSD-IssueNote1-mission-oriented-policies.pdf
[xvii] I wrote several years ago on the challenges in turning the rhetorical commitment to missions into results, focusing on the need to link R&D to regulation, policy, public engagement and other potential blindspots including measuring success: https://www.nesta.org.uk/blog/mission-oriented-innovation-seven-questions-search-better-answers/. See also: Larrue, P., 2021, The design and implementation of mission-oriented innovation policies, A new systemic policy approach to address societal challenges, OECD Science, Technology and Industry, Policy Papers, No. 100
[xviii] Geels, F. (2004) From sectoral systems of innovation to socio-technical systems, Insights about dynamics and change from sociology and institutional theory; Kemp, R. and Loorbach D., (2006)
Transition management: a reflective governance approach; Geels, F. and R Kemp, (2007) Dynamics in
Socio-Technical Systems, Typology of Change Processes and Contrasting Case Studies.
[xix] Jacob, K., Paulick-Thiel, C., Teebken, J., Veit, S. & Singer-Brodowski, M. (2021). Change from Within: Exploring Transformative Literacy in Public Administrations to Foster Sustainability Transitions. Sustainability, 13, 4698. doi.org/10.3390/su13094698; Braams, R.B., Wesseling, J.H., Meijer, A.J. & Hekkert, M.P. (2022). Legitimizing transformative government: Aligning essential government tasks from transition literature with normative arguments about legitimacy from Public Administration traditions. Environmental Innovation and Societal Transitions, 39, 191-205, https://doi.org/10.1016/j.eist.2021.04.004; Braams, R.B., Wesseling, J.H., Meijer, A.J. & Hekkert, M.P. (2022). Understanding why civil servants are reluctant to carry out transition tasks. Science and Public Policy, scac037, https://doi.org/10.1093/scipol/scac037
[xx] The UK government’s funds for using AI for education and precarious workers are interesting, though relatively unusual, examples. These were funded by UK government and managed through Nesta: https://www.nesta.org.uk/project/careertech-challenge/ and https://www.nesta.org.uk/project/edtech-innovation-fund/
[xxi] 13 See: From Science Vision to National Research Agenda in 365 days, (in Dutch: Van
Wetenschapsvisie naar Nationale Wetenschapsagenda in 365 dagen), by Task Force National Science Agenda.
[xxii] Bailar and Smith: https://www.nejm.org/doi/full/10.1056/NEJM198605083141905
[xxiii] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8749321/
[xxiv] https://www8.cao.go.jp/cstp/english/moonshot/top.html
[xxv] Nesta, Experimenter’s Inventory: https://www.nesta.org.uk/report/experimenters-inventory/
[xxvi] Iris Wanzenböck, Joeri H Wesseling, Koen Frenken, Marko P Hekkert, K Matthias Weber, A framework for mission-oriented innovation policy: Alternative pathways through the problem–solution space, Science and Public Policy, Volume 47, Issue 4, August 2020, Pages 474–489, https://doi.org/10.1093/scipol/scaa027
[xxvii] https://media.nesta.org.uk/documents/fundingoptions_.pdf
[xxviii] https://www.vinnova.se/en/m/strategic-innovation-programmes/
[xxix] https://www.oecd.org/greengrowth/2022GGSD-IssueNote1-mission-oriented-policies.pdf p. 23
[xxx] https://www.instituteforgovernment.org.uk/sites/default/files/publications/net-zero-government-climate-change-target.pdf
[xxxi] I recently did a survey of the potential sources and tools: https://covidandsociety.com/net-zero-mobilising-knowledge-easier-effective-decision-making/
[xxxii] See https://www.nesta.org.uk/blog/celebrating-five-years-of-the-uk-what-works-centres/ and Borgman, Christine L., Paul N. Edwards, Steven J. Jackson, Melissa K. Chalmers, Geoffrey C. Bowker, David Ribes, Matt Burton, and Scout Calvert. 2013. “Knowledge Infrastructures: Intellectual Frameworks and Research Challenges.” (May). https://escholarship.org/uc/item/2mt6j2mh
[xxxiii] https://theippo.co.uk/net-zero-mobilising-knowledge-easier-effective-decision-making/
[xxxiv] https://demoshelsinki.fi/wp-content/uploads/2020/12/dh_steering-through-capability2020.pdf
[xxxv] Trondal, J. (2022). An organisational approach to meta-governance: structuring reforms through organisational (re-)engineering. Policy & Politics, 50(2), 139-159. https://bristoluniversitypressdigital.com/view/journals/pp/50/2/article-p139.xml; Trondal, J., Keast, R., Noble, D. & Pinheiro, R. (Eds.). (2022). Governing Complexity in Times of Turbulence. Cheltenham, UK: Edward Elgar Publishing. https://www.e-elgar.com/shop/gbp/governing-complexity-in-times-of-turbulence-9781800889644.html