3. National public sector innovation systems

This chapter examines the reasons why an innovation “shortfall” might exist within the public sector and the consequent need for a deliberate and systemic approach to public sector innovation. It introduces the innovation determinants model which provides a framework for ensuring sufficient innovation, and the innovation facets model which provides a framework for ensuring a suitable mix of innovative activity. Finally, the chapter explores the need for system stewardship.


Brazil’s historical innovation journey, as explored in the previous chapter, describes a country where increased attention has resulted in concerted efforts towards innovation. The journey suggests a growing need for innovation and an evolving awareness that more can (and should) be done to foster, encourage and support an innovative public service. In the context of a clear national need for greater innovation (see Chapter 1), the situation appears ripe for a more coherent approach to public sector innovation in Brazil.

This chapter outlines some of the reasons behind the persistent shortfall in innovation and the consequent need for a deliberate approach to public sector innovation. It introduces models to guide a systemic approach and explores what these might look like in practice.

What might cause a public sector innovation shortfall?

Chapter 1 has identified a gap between the current state and what is desired. Chapter 2 has outlined the historical journey and the case that reform efforts have struggled to achieve their ambitions. Why might this be the case? What factors might contribute to a “shortfall” in public sector innovation, resulting in a gap between innovation happening and the needs and expectations of citizens for better outcomes?

This situation can be partly attributed to a range of “defaults” or inherent biases that exist within the public sector, which can constrain any reform or innovation agenda. These include:

  • Stability and dependability. The public sector has a responsibility to be reliable. This arises from its duty to protect vulnerable population segments (e.g. children, the elderly, and the disadvantaged), safeguard particular functions vital to society and the economy (e.g. the legal system or the financial system), and protect national and intergenerational interests. Accordingly, the public sector often (but not always) needs to be relatively cautious when introducing changes, innovative or otherwise.

  • Risk aversion. This societal safeguard requires democratic governance systems to have accountability, audit and transparency measures in place to ensure the proper workings of government and the public sector administration including support institutions (OECD, 2017). Such mechanisms help ensure public integrity and address core concerns related to effective risk management and safeguarding key institutions. However, such mechanisms traditionally draw attention to failures or (perceived) inappropriateness, even while identifying successful and less successful practices. Defects in the system are of particular concern, whereas positive results are, to some extent, simply expected. This tendency to focus on the negative is often exacerbated in adversarial political environments, where mistakes can be used for political advantage, and encourages a degree of risk aversion in the system.

  • Feedback loops. Similarly, public sector feedback loops tend to focus on avoiding the negative, rather than concentrating on the positive. Positive results are often simply expected or quickly become the norm, whereas negative results become embedded in the institutional memory as lessons on what to avoid. This tendency has driven the widespread development of public sector awards for excellence and innovation, as a means to counteract the focus on the negative.

  • Organisational divisions and operational separation. Such feedback loops tend to entrench a focus on explicit accountabilities. Organisations with clear responsibilities, and the resources necessary to support them, are likely to innovate on a range of fronts in order to meet their obligations. There are numerous examples of public sector organisations that meet this description (e.g. see OECD, 2019a) and are achieving success in this regard. However, many situations where innovation might be required or desired fall into the “white space” between organisations, where ownership is unclear or responsibilities cut across organisations, resulting in multiple potential owners. The need for innovation in such situations may not be met, or it may be hard for those advocating innovation to gain traction against more readily understood and agreed agendas. Yet this is exactly the domain of systemic innovation – novel initiatives relating to cross-organisational needs and demands.

  • Complexity of public sector challenges. This “white space” is a consequence of the fundamental complexity of many of the issues dealt with by the public sector. Such issues are much harder to effectively address with innovative responses, due to the need to meet competing demands and balance different forces in tension with each other. Long-lasting innovation in response to such matters is unlikely to come from short-term fixes or fast timeframes. Rather, it requires sustained investment, deep understanding, a developed ecosystem of partners and a range of interventions over time.

This is not to say that governments and the public sectors underpinning them cannot or do not innovate. Indeed, a fundamental purpose of government is to change things, to steward society from one state of affairs to another. The purpose of outlining these defaults is to recognise that, all other things being equal, there are systemic biases in the way that democratic governance systems currently operate that lean against innovation. In exceptional circumstances – such as a revolution or a crisis situation – these biases may be more easily overcome or even abandoned; however on a day-to-day basis the regular workings of the public sector are weighted against innovation.

Contributing to the impact of these biases is a reality that the “doing of” innovation is simply difficult (OECD, 2018c). This is because innovation:

  • involves challenging the status quo, and all the associated existing procedures, interests and investments

  • is continually changing, as what constitutes innovation shifts, builds upon and transcends what has gone before, and thus always involves new challenges

  • is multi-faceted and multi-natured, involving different purposes, different processes, different skills and different mind-sets

  • is uncertain, as it is something that has not been done before in that context and therefore carries no guarantees as to whether it will succeed, for how long or to what extent

  • introduces change that will often instigate or require further change or adaptation, and is thus uncontrollable and, as such, is in tension with hierarchical and bureaucratic structures

  • has immediate impacts and long-term effects, which may differ completely and be hard to assess, as previous measurements were, by definition, developed for a pre-existing state of affairs, making innovation a difficult activity to cost/or value.

All these characteristics contribute to an overarching innovation shortfall in the public sector.

How might a public sector innovation shortfall be addressed?

Analysis of the Public Service of Brazil highlights both systemic defaults or biases that might hold back or limit public sector innovation, and systemic drivers and contextual needs that require more or greater innovation (as discussed in Chapter 1).

That these two conditions coexist suggests that drivers – the forces pushing for more innovation – alone are not sufficient to spur the needed or expected innovation. A need for innovation is no guarantee of innovation, or the right form or type of innovation. Additional interventions are therefore required.

Innovation must be seen as an ongoing capability

New innovative initiatives do not come out of nowhere; rather, they build on previous experiences, lessons, resources and investments, as well as prior knowledge, technologies and capabilities. This means that innovation is a difficult thing to manipulate or to provide “on-demand”, as previous decisions and efforts matter and shape what is possible. Reactive innovation, occurring in response to specific triggers such as a crisis, certainly takes place, but is severely constrained by pre-existing capabilities. Therefore, the first component of any additional intervention required to meet an innovation shortfall is to treat innovation as an ongoing capability, rather than as a resource to be drawn upon without prior effort or acknowledgement.

Even if some support does exist for innovation as an ongoing capability, it may be treated as specialised or seen as the remit of particular functions (e.g. an innovation lab). However, in a fast-changing, interconnected and interdependent world, the need for innovation may arise anywhere without notice; it may affect a frontline service, the policy development process or an operational support area. Even then, it may be possible to address the situation in a traditional manner by empowering people, providing resources and capabilities, or by bringing in or leveraging external expertise or capabilities.

As society often expects government and public institutions to safeguard and protect their interests (and sometimes anticipate difficulties), it may be unwise to rely on a reactive approach to innovation. This is because the options may be limited in such an eventuality. For instance, they are likely to revolve around:

  • increasing the necessary resources in order to continue delivering existing strategies

  • outsourcing or procuring external capabilities to compensate for any deficiencies

  • reacting to a temporary crisis by unlocking or accessing new resources, capabilities and approaches in an attempt to address the issue at hand

  • overseeing a systemic failure characterised by an ongoing inability to address the issue or shift perspective, despite extra resources that might be made available.

Therefore, it will often not be sufficient to invest only once there is clear evidence that the status quo is unsatisfactory and that a new approach is necessary. Such a strategy might be expensive and demanding without guaranteeing the availability of options and desirable choices, or be otherwise reliant on luck and having made the right choices beforehand.

In short, in order for the public sector to successfully provide innovation when needed or wanted, the underlying capabilities must be invested in and maintained, even when innovation may not be explicitly sought or desired.

A deliberate and self-sustaining approach to innovation

Ensuring that innovation is an ongoing capability requires a deliberate approach. In the absence of such an approach, innovation – because of systemic biases – will be primarily reactive (with the costs that this approach entails). Moreover, barriers or requirements for innovation (e.g. a procurement issue or a skills gap) will likely be addressed on an ad hoc basis, rather than in a holistic or systemic fashion.

While reform agendas and plans, whether in Brazil or elsewhere, often attempt to provide more comprehensive platforms for change, failure to address the underlying systemic factors may result in limited effects.

This is complicated for innovation because supporting and enabling innovation is a continual process of discovery. Each adjustment or response put in place in order to enable further innovation, whether it be about changing procurement requirement, amending legislation or boosting leadership, will reveal new limiting factors or new trade-offs that need to be considered. For instance, if the barrier to innovation is a lack of skilled leadership, and then there is an adjustment to improve leadership competencies and capabilities, it is likely that a new issue will become apparent as the thing that is holding innovation back, such as the need for organisational structures and processes to support innovation. In this way, reform can become a never-ending effort, involving a continual series of change efforts with each simply creating the need for further changes and further decision-making and dedicated effort. Therefore, in the absence of a systemic approach, it is likely there will be a continuing need for intervention. A symptom-by-symptom response will demand ongoing political and senior leadership attention and effort, effort that could otherwise have been focussed on delivering for citizens, and will thus be unsustainable.

If a deliberate and ongoing approach to innovation by the public sector is necessary to better address societal needs, is there already an emerging consensus within Brazil as to what form such an approach could or should take?

Existing knowledge about the innovation system of the Public Service of Brazil

“Regarding the Brazilian literature, although innovation has recently become a trending topic, few relevant studies have been undertaking the challenge of mapping its determinants.”(Cavalcante & Camões, 2017: 19)  

While most of the existing research has been heavily case-based, rather than examining the underlying drivers of innovation, much can be learnt from this analysis.

Perhaps of most relevance is the meta-analysis of Cavalcante and Camões (2017: 30) which highlights recurrent factors from successful innovations. These include the relative advantage of the innovation (over existing interventions), the cost-benefit, the availability of slack resources to introduce the innovation, and network/co-operation and leadership.

The authors also note that public sector agencies tend to generate or adopt innovations “in response to the constant economic, political, social and technological changes in a more globalized and networked world, constrained by rising citizen expectations, complex problems and tight budgets” (Cavalcante and Camões, 2017: 7).

Figure 3.1 identifies some of the key factors influencing innovation at the environmental, organisational and individual levels in relation to the characteristics of the innovation.

Figure 3.1. Levels and influential factors for public sector innovation
Figure 3.1. Levels and influential factors for public sector innovation

Source: ENAP, from Cavalcante & Camões, 2017

These are all helpful insights into the process of innovation and supporting the innovation process. However, without dismissing important existing lessons from history, practice and research, the question remains open as to which levers can easily be manipulated by, or within, government to drive public sector innovation at a systemic level.

“Overall, our results suggest that there is not so much a lack of innovation in the public service as a lack of adequate theories to explain such innovation and of methods and metrics to measure it. Our finding provides strong support for the view that public services innovation does exist, but that its characteristics, determinants and consequences are not the same as they are for either innovation in industry or innovation in services in the private sector.” (de Moreas Sousa et al., 2015: 472)

The innovation determinants model

If a deliberate and systemic approach to public sector innovation is needed, and existing practice within the public sector of Brazil does not appear to provide it, then what form might such an approach take?

In its inaugural public sector innovation system study of the Public Service of Canada, the OECD (2018c) introduced a new model for understanding the underlying determinants of innovation. This “determinants model” provides a framework for understanding the forces that shape whether and to what extent innovation occurs and, in turn, how those forces may be influenced. The model thus contributes to building a public sector innovation system that can consistently and reliably develop and deliver innovative solutions that contribute to the achievement of the goals and priorities of the government and its citizens.

A multi-layered perspective

The determinants model distinguishes between innovation activity at three different levels (OECD, 2018c).

  • The individual. Individuals, on their own or with others, can undertake innovation activity. Often individuals are better attuned to changes in the environment or new possibilities or issues arising where innovation may be needed, and are better able to shift their perspectives than organisations. Innovation at this level will often be, to some extent, personally driven, and will require individuals to draw upon their own energy, resources and time. Such innovation will often focus on specific projects.

  • The organisational. Organisations often have a range of innovative initiatives underway simultaneously. They can marshal resources and co-ordinate efforts in ways that individuals only rarely can, and are generally centred around specific purposes such as health or education. Innovation at this level refers to the orchestration of innovative approaches across multiple people and groups, albeit from a particular sectoral or siloed perspective. As such, innovation at this level is more a matter of process.

  • The system. The public sector as a whole involves multiple government agencies and interactions with citizens and businesses. Innovation at this level relates to meeting collective aims and needs by ensuring diverse activity involving different parts of the ecosystem. Activity at this level is not necessarily about directing activity towards specific aims, but rather ensuring that the aggregate impact of all of the parts addresses societal needs. Innovation at this level relates to collective needs and ambitions.

These three differing levels of the model clarify how innovation as a process and activity plays out differently depending on the scale at which it occurs. Innovation undertaken by a single person is very different to that conducted at a whole-of-country level.

Additionally, these frames of reference can help identify where the focus for innovation activity may lie within a public sector.

For instance, in the case of a formalised system of public sector innovation (where innovation is embedded within the processes, structures, knowledge and operations of the public sector), innovation will likely be a relatively routine activity that is supported as part of the normal way that things are done.

However, if the innovation system is not explicit, and innovation occurs as a by-product of other systems or as a matter of chance, then the focus of innovation will shift to the organisation. In other words, innovation will be driven by particular organisations, either in reaction to crises or as a means of pursuing particular organisational priorities. In such cases, innovation is inherently shaped by a siloed perspective, rather than being driven by collective aims or needs.

In the cases where there is not even a deliberate approach to innovation by organisations, then the focus of innovation will fall to the individual. While individuals are likely to be better able to detect and act upon shifts in the environment, their ability to meaningfully address those shifts will be limited, and reliant upon their ability to convince others. In such cases, individuals will often have to go above and beyond their formal role, dedicating exceptional effort in order to make change happen. A reliance upon individuals being prepared to go above and beyond is a reliance upon luck, as it depends upon the right person being in the right place at the right time. This is the opposite of a deliberate and systemic approach to innovation.

Core determinants of innovation and their manifestations at the different levels

The determinants model outlines four core determinants of innovation – the factors that affect whether and how innovation occurs:

  • Reason. Innovation acts against the status quo and incumbent options, which are usually well entrenched or established. Therefore, there needs to be a reason for innovation to occur, whether it be a specific problem or opportunity, a crisis, a disruption, or a change in context.

  • Possibility. Even if there is a reason for innovation to happen, this does not mean that it will be possible, as existing practices and interventions carry a degree of inertia, and are thus hard to shift to create room for the new. Therefore, there needs to be an opening, a possibility for innovation to occur, such as new resources being made available, or existing resources being removed and thereby making the current option unfeasible. There could also be a political commitment to try something new.

  • Capability. While the possibility of innovation happening may exist, this does not mean that the necessary capability is available, as innovation involves doing something new or doing something differently. Therefore, the capability for innovation needs to be present, whether in the form of technological options, expertise or skills, the processes and structures necessary to enable or support it, or the relevant systems and infrastructure.

  • Experience. While the capability for innovation might exist, this does not mean that it will automatically succeed or become embedded into, or integrated with, other programmes, policies and processes. There needs to be some form of reinforcement or payoff. Therefore, a positive experience of innovation helps to reinforce innovation, whether that experience takes the form of feedback loops that help support the relevant innovation, insight into how to improve the innovation, or progress in scaling an innovation.

The interactions between the different levels of analysis (individual, organisational and system) and the core determinants (reason, possibility, capability and experience) are detailed in Table 3.1 (OECD, 2018c).

Table 3.1. Public sector innovation: determinants of innovation at the individual, organisational and system levels

What level of analysis is the focus?

Core determinants of innovation

Individual (individual effort)

Organisation (collective/shared effort)

System (intersection and aggregate of multiple efforts)

Reason for innovation:

What is driving the intent to innovate?

Motivation to innovate

(e.g. I need or want to solve a problem/try something new; stand out from the crowd/differentiate myself from others; innovate as part of my job/role/identity)

Problem identification/Ideas generated

(e.g. we have to or want to fix policy or service delivery challenges; respond to crises or political priorities; meet stakeholder expectations; achieve/work towards a mission)

Clarity about innovation

(e.g. it is clear that things need to change; innovation makes sense and is a priority; innovation is a responsibility)

Possibility of innovation:

What affects the likelihood of innovation being attempted?

Opportunity to innovate

(e.g. I can work on a project where innovation is appropriate or wanted; apply a new technique or approach in my work; try, or be exposed to, something different)

Ideas generated/Proposals developed

(e.g. we have identified options for how we might solve a problem in an innovative way; processes are open to doing things differently; approval exists to try something differently where appropriate)

Parity of innovation

(e.g. default settings can either be challenged, or are open to doing things differently; system-wide rules or processes do not unfairly bias against innovative proposals)

Capability for innovation:

What is needed to carry out the attempt at innovation?

Ability to innovate

(e.g. I have the tools, skills and resources to undertake the innovation)

Project implementation

(e.g. we have what we need to undertake the project(s), including the relevant skills, systems, technologies and resources)

Suitability for innovation

(e.g. the infrastructure, investment, and commitment exist to ensure that general systems are suitable for innovative endeavours, even when they may not have been anticipated)

Experience of innovation:

What affects whether innovation continues?

Learning from innovation

(e.g. I learn about how people and things react; what it is possible to achieve and whether that change is valued)

Evaluation/Lessons diffused

(e.g. we know what effects occurred as a result of the innovation and have used this knowledge to inform other projects; this helps shape the organisation’s culture and attitude about innovation)

Normality around innovation

(e.g. innovation is not seen as an aberration, an oddity or a frolic, but rather as something that is integrated and built upon)

What the determinants look like at the system level

Given the focus on systemic innovation, it is necessary to pay particular attention to how the determinants manifest themselves at the system level. Table 3.2 illustrates this by:

  • highlighting the relevant characteristic of innovation that underpin the relevant manifestation (i.e. why is it that the determinant reveals itself in that particular way at the system level)

  • explaining the system determinant (i.e. what this implies about a systemic approach)

  • using the example of a crisis situation, which can often spur innovation, as a device to illustrate the determinant (i.e. highlighting the determinant in a situation where innovation tends to occur more easily)

  • showing how the determinants are interlinked and reinforce each other (i.e. why each determinant shapes its successor and is dependent upon its predecessor).

Table 3.2. Understanding the determinants of public sector innovation at the system level

Characteristic of innovation

What this implies for the systemic support of innovation

Illustration of the determinant in a crisis situation (e.g. responding to a humanitarian disaster)

How each determinant builds upon the previous one

Innovation is ambiguous and uncertain Innovation is inherently unclear, and so, all other things being equal, it will come second to any other agenda that is more clearly understood, communicated or measurable.


Additional clarity is required in order to help innovation be seen as a priority. This clarity needs to cover the role, importance and purpose of innovation, and how innovation fits with everything else.

In a crisis there is a clear understanding that things must change quickly, an associated expectation that all the relevant players will play a role in whatever way they can, and consensus that responding is necessary or unavoidable.

If there is a clear sense of why innovation matters, it is more likely that innovation will be given equal attention and consideration when deciding how to proceed.

Innovation is counter to the status-quo

Innovation is new and so, all other things being equal, it will be at a disadvantage in bureaucratic decision making processes that will favour what is known and has already been invested in.


Innovative options need to be given extra weighting in decision making in order to discount or mitigate the bias towards status quo options that have already benefited from investment, time and learning.

In a crisis the status quo is unacceptable, and therefore existing options are unlikely to seem the best or most appropriate choices. New ideas are welcomed, as long as they help meet the present need, and senior leaders prioritise responses.

If innovation is granted attention and consideration equal to that given to status quo options, investments and allocations of resources and priorities are more likely to be oriented towards future needs, and underlying systems will be better suited to new possibilities.

Innovation requires new capabilities

Innovation is about doing new things or doing things in new ways and so, all other things being equal, it will not be as well supported by core operations as existing activities.


The bias of core operating systems towards existing practices needs to be counterbalanced if new capabilities are to be explored, tested and developed before they potentially become the new core operating systems.

In a crisis existing separations of responsibilities may be blurred as resources, people, skills and capabilities are brought in from across the system and new methods are tried, as it is acknowledged that existing ones may not be sufficient.

If underlying systems are better suited to new possibilities, the realisation and enactment of new possibilities will not be deemed as unusual or as costly, and innovation will be more easily integrated into core practices.

Innovation is unusual

Innovation is unlike current practice and so, all other things being equal, it will not be considered normal, unlike processes that are already underway.


Innovation must be actively integrated and linked with core business in order to overcome the default normality of the existing culture and associated behaviours and practices related to current ways of doing things.

In a crisis many habitual and formal processes and expectations are weakened or abandoned, as reliance on them in the face of a disaster is likely to run counter to need. Mistakes are tolerated and even defended as long as they are appropriate to the context.

If innovation is integrated into core practices, it is likely that there will be a much greater understanding of innovation, and why it is important and how it can contribute.

Source: Adapted from OECD, 2018c

The determinants in practice

While each of these determinants can provide an insight into the underlying forces shaping whether and to what extent innovation occurs, the question remains as to what this might mean on a practical level, and how each determinant might be influenced in a deliberate fashion. The following section (drawing on OECD, 2018c) describes some key points of intervention for each determinant.

Clarity – “the quality of being certain or definite”

Are system actors receiving a clear signal about innovation and how it fits with other priorities?

Practical elements of clarity include whether the actors within the system:

  • understand what innovation means from talking about it, seeing it or experiencing it first-hand

  • know why, when and how innovation is a priority, and can situate it in relation to other priorities

  • know how (if) they can contribute to innovation and what role others play

  • see how innovation fits with their shared history and their own context.

Some guiding questions to help illustrate the degree to which clarity exists include:

  • What signal is being sent about innovation?

  • What story is being told about innovation?

  • Is there sufficient clarity about innovation, its value, and what is expected, to ensure that it is a focus?

It is important to note that too much clarity is likely to be as detrimental to the performance of the public sector innovation system as too little (see Table 3.3)

Table 3.3. Finding a balance between too little and too much clarity

Too little


Too much

System actors question why they should participate or engage with doing things differently (“that isn’t my job”).

Actors have a broad sense of what innovation means in the context of the system, how it fits, why it is needed, what their role and that of others is, and know what innovation looks like in practice.

A prescriptive certainty around innovation removes much of the ambiguity, tension, negotiation, and push/pull of innovation, and thus actually inhibits innovation (the asking of “what if?”).

Parity – “the state or condition of being equal”

Does innovation have equal standing with other considerations when it comes to proposed courses of action?

In practice, parity would depend on the degree to which:

  • processes are open to challenge

  • information and decision-making bottlenecks can be circumvented

  • it is easy to find and build a coalition of the willing around shared issues

  • different types of risk can be distinguished, and the difference between risk and uncertainty is appreciated.

Some guiding questions to help consider the extent to which parity exists between innovation and the status quo include:

  • Are existing processes and practices inimical or open to doing things differently?

  • Do those putting forward new ideas feel challenged to improve and develop their innovative proposals through their interaction with the relevant processes and practices, or do they feel exasperated, exhausted and worn down by them?

  • Do those in middle management positions feel able, equipped, empowered and ready to engage with new ideas and innovative possibilities?

  • Do the relevant processes and practices encourage a healthy engagement with risk? Do they generally encourage consideration of the risks of not innovating?

Again, a situation where it is too easy to consider new alternatives may be as harmful as not being able to consider different options (see Table 3.4).

Table 3.4. Finding a balance between too little and too much parity

Too little


Too much

System actors question why they should give any attention to new ideas (“that’s not how we do things”).

Business-as-usual options are not automatically deferred to, but nor is every idea seen as equally meritorious.

Every idea is considered to have equal merit, even where established practice is performing well, leaving decision making and prioritisation bogged down in process and debate.

Suitability – “the quality of being right or appropriate for a particular person, purpose, or situation”

Are the capabilities, systems and infrastructure appropriate and sufficient for the available options?

In an operational sense, the degree of suitability could be influenced by the extent to which:

  • learning occurs from areas that are already matching the external rate of change

  • technologies and their implications are socialised in government

  • new operational models are engaged with and tested and tried in government

  • changing expectations are understood, and any trends and signals that existing capabilities are insufficient are identified.

Some key questions to guide thinking about suitability include:

  • Are the underlying systems of government seen as being calibrated for innovation?

  • How are new technologies socialised and introduced into government?

  • Is there a clear understanding of citizen expectations of government in an environment of high rates of external change and innovation?

  • Does the commitment to and investment in innovation match the rhetoric about the need for innovation?

As with clarity and parity, too much suitability will be as bad as too little (Table 3.5).

Table 3.5. Finding a balance between too little and too much suitability

Too little


Too much

The ability of government to engage with new issues and technologies is likely to be limited as it will not have the requisite understanding or experience, and this will hamper its relevance, appropriateness and effectiveness.

Emergent needs are identified, considered and monitored to track their implications, and system actors have a clear sense of where and when to make investments and commitments so that they will fit their context.

System actors invest too much too early for initiatives that are still going to require significant development/are speculative, or have numerous systems still in development/being refined despite citizen expectations of stability/ consistency of service/experience (e.g. the “bleeding edge”).

Normality – “the condition of being normal; the state of being usual, typical, or expected”

Is innovation seen as integral, rather than as an occasionally accepted deviation from the norm?

Normality in practice might mean:

  • Identifying the behaviours to support innovation

  • reinforcing the links between innovation and regular business

  • socialising innovation

  • upholding innovation.

Some key questions to guide thinking about normality include:

  • Is there a set of identified and demonstrated behaviours for supporting innovation?

  • Is innovation (and the associated breaking with convention/questioning of current practices) valued in regard to career progression?

  • What happens in the event of public criticism of something seen (mistakenly or otherwise) as being innovative? Is the default response defensive or openness?

Table 3.6 considers the right ‘balance’ of normality.

Table 3.6. Finding a balance between too little and too much normality

Too little


Too much

Innovation is seen as a frolic, as something that is not serious or really supported, and will be marginalised instead of integrated.

Innovation is seen as integral to achieving the best outcomes, and default behaviours are supportive of innovation.

Optimisation and efficiency may suffer due to tension between new and existing options. Stakeholders may disengage or become alienated, and change exhaustion may occur.

Ensuring that innovation not only occurs, but that it occurs in a suitable mix

The determinants model and the associated framework for action provide some concrete steps and elements with which to appraise the performance of the system and consider opportunities for innovation. However, while the determinants model allows for insight into whether, and to what extent, innovation is likely to occur within the public sector, it does not necessarily divulge much about the nature of the innovation that might arise. While the system may produce innovation, what guarantee is there that the innovation in question will be socially desirable and suitable to the goals at hand or those yet to emerge?

For instance, innovative activity can be undertaken for a range of purposes, each of which will shape the resultant innovation outcomes (OECD, 2018d). These purposes include

  • enhancement-oriented innovation, where the focus is on upgrading practices, achieving efficiencies and better results, and building on existing structures

  • mission-oriented innovation, where there is a clear goal to be achieved, requiring new approaches and responses

  • adaptive innovation, which focuses on responding to a changing environment with new attempts

  • anticipatory innovation, which is about engaging with new shifts before they become established.

This multi-faceted nature of innovation is presented in Figure 3.2.

Figure 3.2. OECD Public Sector Innovation Facets Model
Figure 3.2. OECD Public Sector Innovation Facets Model

The facets model can help consider the mix of innovation activity taking place and whether there is a sufficient diversity of approaches. Given that innovation is inherently uncertain with no guarantee as to what will succeed, multiple “bets” need to be made to ensure alternative options even if the favoured interventions fail to obtain the desired or needed results. A portfolio is therefore required to reduce risk.

The facets model can also help reflect on the forms of support and investment required by differing forms of innovation activity (OECD, 2019b).

How much innovation?

Even with a framework in place to understand how to act to ensure a more deliberate systemic approach to public sector innovation, the question remains: How “much” innovation is right? If innovation is being performed in pursuit of a range of aims (e.g. increasing efficiency, responding to citizen needs, working towards societal goals or engaging with new emerging technologies) across an entire system, what useful guidance exists to help the public sector assess whether the right “level” of innovation is taking place?

The Public Service of Brazil has clearly been innovating, as illustrated by the Innovation Award and the multiple studies based on the submissions and winners (e.g. see Camões, Severo and Cavalcante, 2018). However, Chapter 1 identifies a range of matters where further innovation would be welcomed or where it may even be essential to the continued successful functioning of the state of Brazil. Economic, social and political concerns all indicate that regardless of the progress thus far, “more” innovation is warranted. In addition, new societal and political priorities, goals and needs will emerge, and unexpected developments and opportunities will arise, each of which will require new ideas, responses and actions to some degree.

However, what “more” innovation might look like is an open question. Three reasons can be given for the difficulty in identifying or quantifying an answer to this question:

  • The impact of any one innovation is hard to quantify.

  • Innovation is a dynamic process.

  • There is no optimal amount of innovation.

The impact of any one innovation is hard to quantify

It is extremely difficult to quantify the amount of change originating from a single innovation, whether it takes the form of the introduction of the smartphone or the creation of an administrative reform that dramatically simplifies how much time someone spends completing their tax assessment. Due to dynamic effects, it is also often impossible to say what the counterfactual would have been in the absence of the innovation. Furthermore, individual innovation initiatives and projects are rarely static in their own right, and the changes they instigate fluctuate over time, whether because of resulting efficiencies or unexpected consequences.

Innovation is a dynamic process

Innovation changes the context in which it occurs. Innovation in one part of the system will, in turn, lead to or require innovative responses in other parts of the system (e.g. see Potts, 2009). Institutions, structures and processes will naturally evolve in relation to innovation, whether they are reacting, preparing or combining with other innovations that occur across the system (e.g. see Juma, 2016). Automobiles changed the market for horses, and transport regulators in turn had to react to and shape the automobile industry. These dynamics created flow-on effects across society, encompassing health, industry and urban planning.

There is no optimal amount of innovation

The “amount” of innovation needed will change continually in tandem with the political, economic, social and environmental context. A major national priority one day may be overtaken the next by an entirely different concern. Equally, a widespread appetite for change may reverse once the costs of that particular direction become clear and the status quo suddenly seems more acceptable. How much is too much or too little will continually evolve as demands and expectations shift.

System stewardship

This inherent uncertainty about the optimal or desirable quantity or quality of innovation points to the need for ongoing stewardship of the public sector innovation system. If there is no inherently optimal amount or type of innovation, the system cannot self-optimise to provide the right amount of innovation in the right forms. It requires ongoing active stewardship.

Stewardship of some form or another is likely needed on a range of fronts:

  • The fragmentary nature of innovation. As innovation is a context-driven activity, it will tend to pull in different directions. For instance, an innovation in one city might look different to an innovation in another city, because of the different specificities of their contexts. Over time, this tendency can fragment the broader regional or national system. While a range of processes (e.g. standardisation, budget processes, etc.) have traditionally reconciled this tendency of divergence, as the rate of innovation increases, it is likely that a more concerted approach may be necessary.

  • Whether the mix of activity being undertaken is appropriate, all other things being equal. Structural drivers and characteristics within the system are likely to push innovation activity in certain directions by default, which may not best serve the overall needs.

  • Harvesting and reflecting on core lessons. As different experiments and innovative attempts occur across the system in different contexts, those involved will learn much about how to support innovation to obtain better outcomes. However, these lessons are unlikely to translate easily between different organisations and settings, and the implications might not be easily understood when viewed from the perspective of a specific context. Stewardship can assist in pooling and distilling the importance of those lessons.

It is an open question as to what such stewardship of the public sector innovation system should look like or how it should function in practice. Traditionally, such stewardship has not been necessary to achieve sufficient results. Therefore, there is little evidence yet as to what such a function should look like. In short, because it has not been needed, it has not happened, so it is speculative to say what it would, could or should look like. However, as the requirement for a more deliberate and systemic approach to public sector innovation continues to present itself, so too will the need for an explicit stewardship function of some description.

The need for a deliberate approach

An ongoing deliberate and systemic approach is needed for innovation to ensure that suitable innovative responses can be generated as and when needed, despite any inherent defaults within the public sector that can, rightly, push against or inhibit innovative activity. Such a deliberate approach needs to answer three questions:

  • Is innovation occurring to the extent needed?

  • Is the innovation likely to provide the right mix of options and choices for the context?

  • Is some form of stewardship present to ensure that the innovation system delivers as hoped?

The next chapter explores the lived experience of the Public Service of Brazil in order to assess how the proposed models suit the Brazilian context.


Camões, M.R.D, W.D. Severo and P. Cavalcante (2018), “Inovação na Gestão Pública Federal: 20 anos do Prêmio Inovação”, Inovação No Setor Públicoteoria: Tendências e Casos no Brasil, Instituto de Pesquisa Econômica Aplicada (IPEA) [Institute for Applied Economic Research], Brasilia, http://www.ipea.gov.br/portal/images/stories/PDFs/livros/livros/171002_inovacao_no_setor_publico.pdf.

Cavalcante, P. and M.R.D. Camões (2017), “Public innovation in Brazil: An overview of its types, results and drivers”, Discussion Paper, Instituto de Pesquisa Econômica Aplicada (IPEA) [Institute for Applied Economic Research], Brasilia

de Moreas Sousa, M., V.dR.S. Ferreira, E. Najberg, and J.J. Medeiros (2015), “Portraying innovation in the public service of Brazil: Frameworks, systematization and characterization”, Revista de Administração, Vol.50/4, São Paulo, http://www.scielo.br/scielo.php?script=sci_arttext&pid=S0080-21072015000400460.

Juma, C. (2016), Innovation and its Enemies: Why People Resist New Technologies, Oxford University Press, Oxford.

OECD (2019a), Embracing Innovation in Government: Global Trends 2019, OECD Publishing, Paris.

OECD (2019b), “Innovation facets: different tools for different aims”, Observatory of Public Sector Innovation blog, 13 March, https://oecd-opsi.org/innovation-facets-different-tools-for-different-aims.

OECD (2018a), OECD Economic Surveys: Brazil 2018, OECD Publishing, Paris, https://dx.doi.org/10.1787/eco_surveys-bra-2018-en.

OECD (2018b), OECD Science, Technology and Innovation Outlook 2018: Adapting to Technological and Societal Disruption, OECD Publishing, Paris, https://doi.org/10.1787/sti_in_outlook-2018-en.

OECD (2018c), The Innovation System of the Public Service of Canada, OECD Publishing, Paris. https://doi.org/10.1787/9789264307735-en.

OECD (2018d), “Innovation is a many splendoured thing”, Observatory of Public Sector Innovation blog, 6 September, https://oecd-opsi.org/innovation-is-a-many-splendoured-thing.

OECD (2017), Government at a Glance 2017, OECD Publishing, Paris, https://doi.org/10.1787/22214399.

Potts, J. (2009), “The innovation deficit in public services: The curious problem of too much efficiency and not enough waste and failure”, Innovation: Management, Policy and Practice, Vol. 11/1, e-Content Management, pp. 33-43.

3. National public sector innovation systems