Working with Change: Systems Approaches to Public Challenges
OPSI workshop, 28 February 2017, OECD Conference Centre, Paris

This workshop aims to better understand how systems approaches can enhance public policies and service delivery through more user-tailored and adaptive governance processes. The workshop is inspired by the work of the Observatory of Public Sector Innovation (OPSI) under the European Union Horizon 2020 framework. Leveraging the experience of both public sector managers and systems thinkers, the workshop will discuss challenges, opportunities and tactics for systems change in the public sector and will define the future agenda for the Observatory work in the field. The morning session was held as a joint session with the OECD’s New Approaches to Economic Challenges (NAEC) initative.

Find below the full agenda for the day – including a video of the morning’s joint seminar with NAEC – as well as photos and tweets from the event, and a recap from a member of the Observatory team. For our main summary and reflections on the event please read our blog post about the event.

The world today is defined by complex and interconnected public policy issues which challenge existing publicservice systems (health, transportation, security, etc.). Large systems designed for a specific purpose (transportsystem to enhance urbanisation, communication, trade) are forced to cope with rapid social and economic changes(sharing economy, environmental and individual mobility issues). Governments are struggling with changingsystems that face technical legacies, silo-ization, neglect, over-burden, or reductive thinking (simplified, linearthinking). Many of the underlying assumptions about how to define and solve public policy problems havestagnated – the result has been the application of old lenses to understand today’s reality and a belief in thecapacity of last century institutions to solve 21st century problems.

Systems approaches attempt to challenge these limitations, opening up public policy systems and re-examiningproblems from a user-oriented perspective. By focusing on the public value connected to specific policy problems,systems approaches can help un-pack and re-orient government decision making away from processesimprovement to focusing on the outcome. For example, we might assume to know what education is inherently for,but do new teaching methods alone enable students to develop the skills and talents that are needed to thrive in aradically new economic and social reality? By starting to understand the new meaning of problems, systemsthinking helps governments to become more adaptive, change public policies in line with new challenges andcreate systems that support renewed or even entirely new forms of public value.

At the same time, systems approaches are not yet widely adopted in government. More downstream, user-focuseddesign tools – looking at simplifying known realities rather than understanding complexity – are getting traction inthe public space, with an associated risk of quick solutionism. Only looking at the user “end” of the public servicescontinuum leaves the overall systems unchanged, with more meaningful impact not achieved. However, changingsystems is hard – institutional legacies, failure to understand complexity and lack of leadership support are some ofthe factors inhibiting change – and government can’t just shut down old systems while working on new ones.

This workshop is intended to develop a better understanding of how systems approaches can enhance publicpolicies and service delivery through more user-tailored and adaptive governance processes. It is inspired by theresults of an OECD report on systemic approaches prepared by the Observatory of Public Sector Innovation (OPSI)under the European Union Horizon 2020 framework. It brings together public managers, systems thinkers andacademic experts to discuss how systems approaches can be used to redesign the architecture of today’s problemsand deliver public value. It will discuss concrete cases where systemic approaches have been applied, the lessonsthat have emerged from design and implementation, and what tools can help public managers initiate a successfulprocess. The expected result of the workshop is to identify the key opportunities and challenges to implementsystemic change in public policies and design a research plan of action for the OECD going forward.

Detailed agenda

Public sector systems are under stretch as they face complex, multi-dimensional and multi-stakeholders problems. Piecemeal reform produced with traditional analytical tools and problem solving methods has achieved a tipping point and is no longer producing results in many areas. Yet governments are still struggling to identify what does not work, to abandon traditional ways to approach policy issues, and to equip themselves with new tools that allow them to work with relative precision in a context of ambiguity and uncertainty. This session consists of a panel bringing together government decision makers and experts to reflect on the need, opportunity and challenges of systems change in the public sector.

A selection of the following topics will be addressed in the panel:
  • What is broken in public sector and why do we need systems change to fix it?
  • Why have systems approaches not been taken up in the public sector?
  • Which complex policy problems would benefit from systems approaches?
  • What kind of change – institutional, cultural, operational etc. – is needed in the public sector to implement systemic change?
  • What is the role of leaders in initiating and stewarding systemic change in the public sector?

The NAEC Seminar was opened with remarks from:

  • Juan Yermo, Deputy Chief of Staff, Office of the Secretary General, OECD
  • Matthew King, Head of Unit, Open Innovation, Directorate-General for Research and Innovation, European Commission

The panel discussion was moderated by Rolf Alter, Director for Public Governance and Territorial Development, OECD, and was introduced by Edwin Lau, Head of Division, Public Sector Reform, Directorate for Public Governance and Territorial Development, OECD. The panellists were:

  • Oli-Pekka Heinonen, Director General, Finnish National Board of Education, Finland
  • Riel Miller, Head of Foresight, UNESCO
  • Sir John Elvidge, Former Permanent Secretary of the Scottish Government, Chairman of Edinburgh Airport Ltd., United Kingdom
  • Chad Hartnell, Director of Operations of the Innovation Hub, Privy Council Office, Canada

Watch the video of the session:

This session will provide workshop participants with the opportunity to learn about systems change approaches and tactics from concrete experiences which have been documented as part of the OECD work. During four parallel break-out sessions moderated by experts, participants will learn about the systems approach journey from those who have done it and go through the process of systems change in the public sector.

  1. Regulating the sharing economy (Canada)
    Joeri van den Steenhoven, Director of MaRS Solutions Lab presented the case study of how the city of Toronto (and by extension other cities across the province of Ontario) developed new regulations for the sharing economy in public transportation. The session was facilitated by Justin Cook, Senior Lead for Strategy at the Finnish Innovation Fund, SITRA and Member of Rhode Island School of Design
  2. United against domestic violence (Iceland)
    Alda Hrönn Jóhannsdóttir, Chief Attorney and Marta Kristín Hreiðarsdóttir, Specialist, from the Reykjavik Metropolitan Police presented the case study of their United Against Domestic Violence programme in Iceland which is introducing a new integrated support systems to victims based on a notion of domestic violence as a social problem that affects everyone, rather than solely as a private problem. The session was facilitated by Marco Steinberg, Founder & CEO Snowcone & Haystack and Advisor to the MIT Collaborative Initiatives.
  3. Experimental Policy Design (Finland)
    Mikko Annala, Head of Govanance Innovation at Demos Helsinki, presented the case study of how Finland is developing a new framework for experimental policy design, combining systems and design thinking approaches to develop a new framework for carrying out experiments in government. The session was facilitated by Piret Tõnurist, Policy Analyst, Lead on Systems Thinking at the Observatory of Public Sector Innovation, OECD.
  4. A purpose-driven child protection system (Netherlands)
    Marc Dinkgreve, Ambassador of Knowledge, Jeugdbescherming Regio Amsterdam (Child and Youth Protection Services in the Amsterdam region), presented the case study of how they have redesigned their approach to child protection around the guiding mission of “Every Child Safe, Forever”. The session was facilitated by Dan Hill, Associate Director / Head of Arup Digital Studio, United Kingdom.
The session will discuss the tactics and approaches to initiate systems change, and how these can be applied in different contexts. The session will be introduced by the presentation of 10 main Principles of Systems Change coming out from the OECD Horizon 2020 work. It will proceed with a moderated group discussion where participants will work in groups to solve a systems change challenge based on the instruments and approaches illustrated during the precedent sessions.

Presentation by Justin Cook on principles of systems change:

This session will bring together the findings from the group discussions and different levels and perspectives on systems change. The aim is to discuss what emerging approaches seem to be working and what challenges lie ahead, and how OECD can contribute to enlarge the knowledge and evidence bases on systemic approaches in the public sector.

A selection of the following topics will be addressed in the panel:
  • What can we do to lower the barriers to system change?
  • How do we avoid shutting down existing systems while changing them?
  • What are the cultures needed for system change? How can governments work with uncertainty?
  • What level of support (leadership, resources, etc.) is needed to drive effective system change?
  • What can the OECD do to broaden the evidence-base of systems approaches and support governments in their efforts to explore the potential of systemic change?

The panel discussion was moderated by Luiz de Mello, Deputy Director, Public Governance and Territorial Development Directorate, OECD. The panellists were:

  • Joeri van den Steenhoven, Director, MaRS Solutions Lab, Canada
  • Marco Steinberg, Founder & CEO Snowcone & Haystack and Advisor to the MIT Collaborative Initiatives
  • Dan Hill, Associate Director / Head of Arup Digital Studio, United Kingdom
  • Monika Queisser, Senior Counsellor, Directorate for Employment, Labour and Social Affairs, OECD
  • Mario Cervantes, Senior Economist, Committee for Scientific and Technological Policy, STI, OECD

Photos and Tweets

You can view all tweets related to the event by searching for #OECDsys

Recap of the day

To provide an overview of the experience of the day and some reflection from the day, Alex Roberts, an innovation specialist in the OPSI team, provides a recap of the day below.

The discussion in the first panel of the day covered a range of topics and issues. The following were some of my takeaways:

  • This quote from John Maynard Keynes (1930) which was shared by one of the speakers: “We are suffering, not from the rheumatics of old age, but from the growing-pains of over-rapid changes, from the painfulness of readjustment between one economic period and another.” The difficulties we currently face may be an issue of transition between one state and another.
  • That there is increasing complexity in the world and that we live in a time of volatility, uncertainty, complexity and ambiguity (VUCA), where the traditional model of policymaking is no longer appropriate. More and more actors are leveraging technology to have an impact on the world, making for more and more complexity.
  • Where once governments and public administrations thought they could solve difficult problems through regulation and the provision of guidance, we are now realising that we need to take integrated action that considers the system as a whole. This is a different role for government, and requires continuous engagement with problems rather than less.
  • This is where systems approaches come in, and as identified in a preliminary report “Working with Change: Systems approaches to public sector challenges” has been used to some success in a range of diverse settings: child protection in the Netherlands, responding to domestic violence in Iceland, engaging with the sharing economy in Canada, and in experimental policy design in Finland.
  • A lot of new societal and regulatory issues are arising out of new technologies and business models (such as the sharing economy and autonomous vehicles), and these require administrations to work across silos and boundaries, and this can be challenging. It requires bringing in new classes of stakeholders into policymaking.
  • This does not mean that the public sector is broken – it is functioning as it always has. Rather, it is the context that has changed, which has meant that the public sector is misplaced. It is part of a system of thinking from 200 years ago, and we need new mental models that can respond with speed, interconnected responses, and that can deal with complexity and uncertainty.
  • Yet more and more citizens view the system as broken. While administrations might take comfort in the notion that most of their systems work for most of the people, most of the time, people are focussing increasingly on the points of failure, of where systems are not working. An education system that fails one in five children can no longer be viewed as “good enough”. These failures are not an issue of the competence of the public sector. They are a matter of cultural change – of public servants not only being able to understand and engage with complexity, but of being able to change themselves, and their processes and systems. It means confronting the question of not only “what needs to change” but “how do we need to change”. Until that happens, changes will not lead to the necessary shifts to get the results that citizens expect.
  • This tension means that systems and complex problems inherently involve talking about power – it is unavoidable. Yet there can be a tendency for public sector responses to complex problems to be strategies that actually reconfirm the existing system, that allow players to stay as they are, rather than changing and evolving as needed for a new system.
  • This behaviour is supported by public administration systems that are layered with risk aversion, and which gravitate towards the most polished proposals, rather than the best ideas which might involve big shifts and uncomfortable changes to thinking and behaviours.
  • Cultural change for the public sector will involve confronting the understanding of what it is to be a professional public servant, and to shift from:
    • An expectation of knowing the answer to being prepared to not know
    • A reliance on planning to being able to recognising when not to plan
    • A clear division of labour and responsibilities to encouraging self-organising
    • An avoidance of errors to a willingness to make use of errors and the lessons that come with them.
  • Such change will not come from people being told what to do; it instead requires giving people the room to experiment, to learn from doing, and acting on those lessons through self-organising. Those lessons in turn will spread better through networks rather than identifying solutions, trying to appropriate the lessons and imposing them by dictate more broadly.
  • Therefore part of the response to a complex world will be for governments to give up a command and control approach to administration, though this is something that is deeply uncomfortable for the public service. No response will be effective unless it looks to the capabilities of the broader society to contribute, to be empowered with the freedom to use and exploit new tools and technologies. Government alone cannot solve the problems, so other parts of the system must be able to contribute. It will require new partnerships and new forms of interaction with non-state actors.
  • This realignment of power, of giving people freedom to act and learn, must be reflected in public sector structures, where the central areas give more freedom to those on the front-line, and give rise to a situation where systems thinking is seen as the norm and linear thinking as the exception. People need to feel empowered to think that spontaneity and improvisation is not a dereliction of duty.
  • Another component of a response is to look at the future not as something to be feared and colonised, where we impose our world view on it and then get stressed when that does not work, but to use it as the basis of different conversations where wicked problems are seen as the fascinating emergence of interconnected systems and an opportunity for the realisation of new possibilities.

It was a very interesting session and there seemed to be a consensus on the need for freeing up administrators from some of the old structures and conceptions of how the public sector works, as well helping public servants confront and reflect on their beliefs about their role and how they work with other parts of the system in a far more collaborative fashion.

In this session participants were split into four groups to respectively learn from some of people involved with the four case studies. After a presentation on the background of the respective case, participants were challenged to think about:

  1. What are the most important factors for (e.g.) changing the system to tackle domestic violence?
  2. What are the possible barriers to systems change?
  3. Additional questions that need to be answered?

In the session I participated in, we looked at the example of Child Protection Services in the Netherlands. It was fascinating to hear about how the system went from a state of crisis to a system focused on the notion of “Every child safe, forever”, and the difficult work involved in making that shift (and then making it stick).

I felt the session was a great chance for people to digest the cases, to think about what was important, the links between different components of the system, and what that means for implementation/intervention. The hands-on nature of the workshop was good for helping think about the case more broadly and what a systems approach can involve.

At the beginning of this session Justin Cook, one of the authors of the report, shared ten points for how participants could think about enacting systems thinking and systems change:

  • The End of Known Knowns – that the world operates by a new set of rules that we understand less and less, and the need to be honest that we actually struggle to understand some of the simple things, let alone the bigger ones.
  • The Complexity Gap – the gap between the problems faced by institutions and their capacity to tackle them. This has led to the rise of systems thinking, and other systems approaches such as design thinking, which can help identify options where change can have the greatest impact.
  • Proximate Failure, Distant Impact – when systems thinking is done, it will result in some kind of noticeable failure, whereas the successes can be longer term and harder to track (this does not fit very comfortably with the traditional public sector approach). Prototyping can help ameliorate some of this failure.
  • Fuzzy Fronts & Open Ends – projects need to be much less certain at the beginning, and also have more open ends in terms of what might happen. (Again, this does not fit comfortably with the traditional public sector idea of project management.)
  • Beware of Toolkits – toolkits by their very nature are reductive, and thus divorced from the principles they are based on. Systems change is hard, it takes time, and it is a craft and an art, rather than a standardised process.
  • Mens et Manus – systems change requires a learning system with strong feedback loops between analysis and execution, that allows learning from doing. This is not easy.
  • Contextual Variance – practitioners should be sceptical of the notion of scaling-up of solutions. The best to hope for is an agreement around a set of shared principles, but then have those considered in regards to specific contexts.
  • If You Know Everyone in the Room, You Are Going to Fail – systems approaches require diversity in people (both inherent and acquired). You need to have the usual players, but also some of the unexpected. Along with prototyping, diversity is one of the hedges against failure.
  • Work Both Hard and Soft Constraints / Manage Expertise – no single expert or domain of expertise should be dominant, expertise should be subordinate to the objective.

Justin advised that the products of systems design should focus on:

  • A vision for an alternate future
  • The principles that govern that future
  • A portfolio of interventions that can be undertaken to help realise that future.

Participants were then again split into four groups and were asked to consider the issue of Ageing in the 2st Century from a systems perspective. Participants were asked to think about:

  1. What is the problem?
  2. What are the opportunities and benefits?
  3. What is our vision for ageing in 2050?
  4. What five actions can be taken to initiate a systems transformation process?

In the second part of the exercise, participants were then asked to visualise the outcomes of the new system in the form of an “OECD Ageing in 2050 Factsheet”.

I found the session a really interesting exercise for teasing out assumptions, connections between different elements of the system, and for helping to re-frame the “issue” as an opportunity for achieving a better state. A big thanks again goes to the facilitators for helping run some really interesting conversations.

In the final session of the day, a closing panel summed up some of the discussions, shared some insights and started the conversation of what might happen next. It’s hard to capture all of what was covered, but some of the things I took from the session included:

  • Where you start with systems change depends on the challenge. You need to look at the capacity of the people in the system, and sometimes you need to start with solutions, sometimes by looking at policies/services, or sometimes with capacity building. Do a quick analysis of the system and then start with something small and see what happens.
  • An approach like that, that starts with projects, can build a growing cadre of people interested in systems change. Work out what you need, then involve the people with those capabilities, then do good work and which those involved will enjoy, and then they will move on to other projects and want to do it again, and involve other people.
  • We need to make it harder to kill innovative projects. We need to change the defaults where permission is required for every step.
  • We underestimate the risk of the status quo, and overestimate the risk of doing new things.
  • Start by starting – build a sense of strategic orientation, of where you are trying to go and what you are trying to achieve, but do not pretend you know all of the steps that will be involved from the beginning.
  • People in the middle layers of organisations tend to be concerned with preserving the organisation – that is their role, and they are behaving exactly as they should be given the way the system is designed. Change therefore will more likely come from the top and the bottom.
  • Scaling of solutions is challenging and is often not appropriate – yet at the same time, we must not be afraid of scale. If you do not design for scale, then you risk losing relevancy. Governments need to consider different ways of getting to scale, including through experimentation.
  • At the same time, experimentation needs to be handled carefully. Something designed as a “pilot” is unlikely to ever go beyond being a pilot.
  • Not every problem is wicked, and systems thinking is hard and can be laborious. Yet as we also learnt from the case studies, sometimes it can achieve amazing things.

The role of the OECD can be to bring its validating power to the art of systems approaches, to bash and test the ideas around it, and look at new ways of convening the networks that might be interested in this topic. Thought needs to be taken to consider the best formats to convey the information contained in the report, to bring it alive for people in a way that leads to systems doing.