Policy Design: Towards Understanding and a Methodological Framework
Design Exploration series, Strategic Innovation Lab, OCADU, March 2017
(Excerpts of the study)
“I’d rather be vaguely right than precisely wrong” (J. M. Keynes)
Summary
Interest in policy design has been increasing, but it is still being addressed by designers and policy researchers in isolation from each other. The design community has not produced much research on policy design and was has been done often lacks references to policy studies. Akin to some business research, most concepts and methodologies regarding policy design in the design community come from design consultancies, which tend to seek business promotion more than solid research. The policy community, while producing much more research on this topic, is divided between those who do not consider policy design a valid notion at all, and those who approach both design and policy in much narrower terms. Moreover, in policy studies we can rarely find references to design research (beyond the so-called “design thinking” hype). Such isolation might be particularly surprising when we realize that policy and design share some of the same roots in complex social systems approaches and decision-making, amongst others.
Why is this important? Firstly, if policy design is to be pursued for systemic social change it needs to be properly conceptualized and operationalized. Secondly, policy design might be a great new opportunity for further development for design research and practice — while also addressing some of the counterproductive tendencies in policy (e.g. techno-economic rationality, “evidence”-based policy, “deliverology”). Thirdly, it is timely to bring back a more comprehensive understanding of design and of policy that has the potential to deal with the predicaments of complexity, stakeholder involvement, and working across silos. Nevertheless, the understanding will not suffice without identifying practical ways to deal with the real-life challenges of designing in the policy space.
After Introduction and the Overview of the approach, the study report proceeds with consolidated representation of fundamentals On Policy and On Design. Different approaches to policy design are interwoven with author’s arguments and critique in the chapter on An Understanding of Policy Design. The proposed new approach to policy design is outlined in the Towards a Methodological Framework. Final thoughts and considerations for further research are presented in the Discussion. Overall, the study report represents an intermediary step towards a more comprehensive and practical approach to policy design.
The study is the result of 15 months of structured research at the MDes in Strategic Foresight and Innovation (SFI) at the OCAD University, and several more years of a less structured personal research. It also builds on two decade long professional experience of the author in the space of social systems innovation, and policy and institutional change across more than 25 countries. It also benefited from consultative support and collaboration with professors Peter Jones, OCAD and professor B. Guy Peters, University of Pittsburg — as well as from peer-learning in the 2015 SFI part-time cohort. Early insights from the study were presented at two webinars for the United Nations Develop Group (for a global fund on integrated policy support), and elements of it were part of the author’s presentation at the panel on Advanced Policy Design at the Relating Systems and Design symposium 2016. The findings of the study were presented at the Explorations in Design at OCAD (March 21, 2017), which contributed with peer critique and led to its finalization.
Introduction
This study aims at addressing an important gap in current body of knowledge: that of conceptualization and application of policy design. There is a growing interest in policy design, but this has been addressed by policy and design communities in almost complete isolation from each other. Rarely do we find references to design literature in policy studies, and there are very few reference to policy literature in the design studies. Hence, the large gap between how designers approach policy and how policy professionals approach design requires a new approach that this study hopes to bridge.
The study focuses on the relationship between two concepts and their related practices: design and (predominantly public) policy. Each of the two has been evolving individually (due to, amongst other, emerging needs for addressing complex social issues), while some of the novel thinking, approaches, and methods are often being introduced by returning to the original propositions in the light of new contexts. Nevertheless, the relationship between policy and design remains problematic — with a tendency for oversimplification on both sides. In design, it is most evident in the trend of codifying models of “design thinking” and extending product design into design in complex settings; and in public policy, there has been a tendency to narrow the whole concept to policy instruments. Thus, whether or not there can be policy design at all, depends on what we mean by design and what we mean by policy. However, to change practices, new understanding needs to be accompanied by new practical applications. No current approach seems to be adequate for dealing with policy design in more comprehensive and effective terms.
It should be noted that the study is not a scientific one: it does not seek to find the “truth” about policy design. Neither it is an artistic representation of policy design. Instead, it is design-based: it tries to bring about a new understanding of policy design so as to induce change in the way it is applied in practice. In other words, the study hopes to contribute to the betterment of policy design by proposing a methodological framework.
It should also be noted from the start that many in both of design and policy communities still have considerable reservations about the very notion of policy design. Some of the challenges in addressing policy design originate in the different understanding of key concepts, but also relate to perplexities of recent developments in both design and in policy.
This study should be considered in terms of its specific understanding of design and of policy, which led to the specific framing of policy design — beyond creative problem solving and beyond design of policy instruments. The arguments made in the study do not support the optimized representations of processes — either of design or of policy — so it does not focus much on superimposing one upon the other, or on identifying which aspects of policy and of design relate to each other. However, it will make references to design processes and methods, and the typical stages of the policy process.
The destination (“desiderata”) of this study was reached in an oblique manner and diverging into apparently not related concepts and practices through a wide research scope. Moreover, it has a “double-edged” purpose (Skocpol, 2003): to interact between theoretical concepts and practical applications. However, in some ways, it ended where it might have begun: with the core definitions of design and policy. As it will be presented later, even basic research of more systemic aspects of design and of policy reveals that their main elements align very well.
The starting point of this study was counterfactual: that there is (or cannot be) something called policy design. The basic premise was not that policies could not be designed because design cannot deal with complexity or uncertainty. Instead, the notion that “design is best when knowledge is scarce” (H. Kerr) was taken as highly valid from the start. As it will be presented in this paper, the kind of approach to design presented here is ideally suited for dealing with the perplexities of policy development. The premise that lead to this counterfactual was not even related to the problems of applying conventional product and service design to certain aspects of policy. Regulations could be treated as “products”, and policies do involve service delivery.
The main issue that led to the counterfactual was that design is often considered to be intentional — to the extent that intentionality in some tasks is often colloquially referred to as being done “by design”. This indicated that the basic validity of the notion of policy design might be undermined from the start. However, as the study will try to argue, the intentionality in both design and in policy is a complicated notion that might not end up being an obstacle, but actually an opportunity for applying design for policy.
The second issue related to culture. Apparently, the cultures of design and of policy are very different, if not completely opposite. This position is mostly based on superficial generalization of what designer and policy professionals represent. Very often designers are portrayed as “creative” people who come up with “crazy”, “innovative” ideas by being “playful”. Policy people are often presented in completely opposite terms. However, the two cultures might be very similar. The core is represented in the notion of judgment (practical wisdom). As it will be presented later there is no stretch to almost completely equate the notion of judgment in policy making with judgment in design[1]. It is exactly the judgment that helps designers and policy professionals deal with ambiguity, complexity, and uncertainty, i.e. to cut through the overwhelming or incomplete knowledge about the “wicked” problems. Furthermore, the main design competencies and attitudes identified in this study could be as aligned with the design as with the policy culture.
We should also recognize that both policy and design are forward-looking. Design is about “that-which-does-not-exist”, while policy is a future intervention in(to) society. Tendency to connect design and policy by scientizing them (not least with increased emphasis on “evidence”) should be counteracted by bringing each of them individually — and both together — into the realm of anticipatory models and strategic foresight. In this space, design and policy would have much more in common that we realize.
The breakthrough in this study took place through the three particular insights. The first was the understanding of design as the 3rd culture of human inquiry (Nelson and Stolterman, 2012), i.e. design as intervention for change (not truth (science) or personal expression (art)). This 3rd culture is exactly where design and policy intersect. The second insight related to the notion of policy in the context of norm-seeking and value configuration (Özbekhan, 1968). And the final insight was based on the alignment between the design “thing” coming out of the broader concept of collective and dasein (Latour, 2008) and the notion of “thing” (as the traditional assembly in proto-democratic German tribes[2]) relating to one of the central arenas for policy.
Towards a Methodological Framework
To be authentic to one of the main underlying themes in this study regarding interactivity in complexity (problems-solutions-instruments)[3] — I would like to acknowledge that the “solution” to the understanding of policy design proposed here has already been reflected in interpretation of policy, design, and policy design earlier in this study. So, there is not anything really new to argue, except to try to articulate it in more systematic manner and for practical purposes. Considerations of risks should be avoided at this stage, because that would lead us to move from an “ought-to” future towards feasibility, and eventually be trapped in extrapolating the past into the future — or, at best, doing the business almost as usual. Instead, we need to be imaging another future of policy design and work on approximating the present to it. This is a balancing act — rather than a final solution — to which we need to continue on a long haul.
I am not proposing some sort of the ultimate solution or a “grand theory” of policy design. Nor am I naïve to think that a more comprehensive policy design could be easily (if at all) applied in practice. However, I do posit that we need to improve our conceptualization and understanding of policy design — other we end up in the trap of “drifting goals”[4]. We should be setting higher standards for what is attempted in real-life situations and be more integrative and authentic in the ways we relate to the narrative we use. We should not be making references to wicked problems or “muddling through[5]” and then end up using optimized procedures to solve linearly such problems (and then colonize the concept of policy design with that approach). Such misalignments are only making scholarship weak, and practical efforts increasingly ineffective. Thus, I advocate for closer relationship between policy and design studies, and more integrated approach in each of those.
…
Key considerations and principles
As mentioned earlier, this study does not focus on processes. To be clear, processes are good for as long as we understand them as maps (that somebody else made) and not the territory[6]. They are good when we deal with (semi)-structured, and tame problem and when we design in the “glass box”. Secondly, simple mapping of a standard design process over a standard policy one does not really produce much insight. Besides not reflecting the real-life, “liquid” reality, both of those processes are based on the generic approach to problem solving — so they do overlap quite nicely. The problem is that is that when we claim, for instance, that the (sub)stage of developing policy alternatives (from the policy process) relates to divergence in design, we do not really say anything. On one side, those two stages might mean very different things in practice (depending on the practice of policy and of design) — sometimes as much as decision attitude differs from design attitude (Boland and Collopy, 2004). On the other side, they could be made rather same, but combining different methods used in each. Nevertheless, these generic processes present the sequences which are typical, but activities in actual reality are less linear — and what really matters is what happens inside those stages and how these related to the broader context Neither of which is reflected in generic policy and design processes — nor it could be made so.
Henceforth, processes will be considered as a heuristic, and a way of communicating ideas — rather than something actually happening in policy design.
Besides the perplexities of intentionality and the need to “dance” with complexity widely discussed in this study earlier, we should consider the need for increased “density” in the policy design space (Carlsson, 2009). We need such “mess” of different perspectives to effectively design with requisite variety and avoid “optimal solutions” to “wickedness”. The dominant approach to policy in terms of framing social problems as structured or semi-structured (Hisschemoller and Hoppe, 1996) is that it limits participation and perspectives[7]. It represents the main obstacle for applying more comprehensive design in policy. In contrast, the strategy of “policy as learning” recognizes the complexity; social problems being “conventions”; the iterative approach to policy; as well as the notion of turning existing situation into “preferred ones”. And it invites diverse stakeholders to come forth, while also sharing the responsibility for the outcomes. Furthermore, that would broaden the notion of policy designer (rather than keep it within the community of professional designers, policy makers, or design labs) to include wider public. While highly challenging — and not without unintended consequences in the age of social media — it would help us use policy design as genuine “lessons in democracy” (Schneider and Ingram, 1997) and reinstate assemblies and parliaments as democratic Agoras.
For that happen, we also need to move away from dominant techno-economic reasoning and embrace political one. We would still need all 5 types of rationality (Diesing, 1973) — including social and normative ones — but the political should always precede and encompass considerations of feasibility and efficiency, and bargaining and economizing. This would focus us more on norm-seeking tasks (Ozbekhan, 1968) and trains us in making good judgment calls (Vickers, 1995; Nelson and Stolterman, 2012) collaboratively.
Political rationality is very “designerly” — beside relating to creativity (as an “organizing principle”) it requires dwelling in a permanent state of tension between differentiation and unification. Such more agonistic context would imply that the more tension (and designerly “agony”) we can tolerate, the more “politically rational” actions we will take. Dealing with complexity by “intelligence” means holding two or more contradictions while being “thrown” into design situation with its own dynamic and path dependences — the only meaningful way to deal with contemporary (social constructs of) human predicaments.
The quest to remain in a “fluid state” and avoid “early closure” should be balanced by the need to have a closure after all. Decisions need to be made and actions need to be taken in real-life situation — and for that we need more of democracy and more of politics, and less of technocracy and managerialism. However, political actors would need to identify ways to “satisfice” and to establish “stopping rules” when dealing with social “wickedness” — as well as taken on the political accountability for such choices.
The focus of policy design should be on social change — and that is intrinsic to both policy and design. Such “betterment” is nurtured by changing behavioural patterns, power relations, and resource flows by manipulating leverage points and value configurations for increasing possibilities (i.e. development). This essentially implies engaging in systemic social innovation — this time government not being the sole, or even the most central, actor in that space. And the higher on Donna Meadows’ scale of entry points for system change we can get[8], the more effective will be the results.
We cannot design values — as we cannot design experiences — but we can design for values. Essentially, such approach would lead us to a norm-seeking grand narratives and co-creating the future as it emerges. However, that would require considerably increasing the horizons of our considerations and the breath of our perspectives — while designing for emergence. But, we should also acknowledge that not all can (or should) be designed and that many policy outcomes would still be produced by pure luck or mere necessity.
The notion of “real design problem” (Nelson and Stolterman, 2012) comes to the rescue. We should not seek problem solving or producing ideal designs, but rather work on approximation of idealized to what the actual. Such deliberate design strategy aligns well with the need to muddle through in policy. To quote John Flach on design and politics of muddling[9]: “In order to make change happen, designers have to be prepared to participate in the muddling through process… designers cannot sit outside the sociotechnical system and throw solutions over the fence… they have to engage with the social dynamic of sensemaking within the organization… negotiate with multiple stakeholders… be satisfied with the incremental changes that typically result from such processes”. Hence, we would need less of prototyping and piloting, and more of probing — or to return once more to Bruno Latour[10] who posits that: “It has never been the case that you first know and then act. You first act tentatively and then begin to know a bit more before attempting again”.
Models of policy design
I envision two basic models of policy design. One represents a bricolage of diverse approaches — another one is fully integrated “generic system” in the form of integrated platform. Both of those are based on comprehensive understanding of policy, design, and policy design, but they lead to different strategies and produce different implications.
…
With regard to methods, both models can explore the wide and diverse variety of methods and techniques from design, dialogue, and policy — as well as incorporate from other fields. In contrast to usefulness of toolkits for particular aspects of policy in the “bricolage” model and toolkits for using platform/s in the “integrated” model, I would argue against a standardized toolkit for policy design overall.
…
Finally, we should emphasize the triad foresight-design-innovation in the context of policy design. Foresight (in particular strategic foresight) should replace the dominance of forecasting in policy — and should also be further integrated into design, despite the proposition that any design is essentially foresight-based. For the purpose on working on policy, typical advanced foresight methods would need to be integrated in what I call “Policy Design from the Future Present[11]”. Because of the long policy cycle, this approach would help prevent reactive or proactive approach to policy design and enable the interactive one. Finally, policy design will be effective to the extent it is capable of innovating: producing the results that matter and “work”.
Core competences for policy design
As emphasized in this study, one of the most important aspects of designing in the policy space related to design attitudes, competences, and the broader culture. It was argued already that the culture of design and that of policy might not have been as different in the past as they are today. But, even today, the differences — when we generalize them — are not so great. Nevertheless, policy space does not incorporate enough of design culture, and any attempt to pursue policy design would need to address it.
The list of possible core competences of policy design is rather long, as presented in the chapter on design in this study. More work is required on consolidating them, but the first attempt is presented below. The map represents 3rd iteration. In the first, a total of 53 core competences were included and they resulted in as many as 22 levels — of those only 11 levels being connected (although the remaining did form several cycles). The second iteration included 28 core competences and also resulted in a complicated map with 15 levels (this time all connected). The final iteration included 16 most important core competences, with 6 levels.
…
Some implications of the methodological framework
In the Gardens of Democracy, Liu and Hanauer (2011) argue that the government should move towards addressing the “Big What” and the “Small How”. While this might be valid to some point, the methodological framework in this study expands the field of such considerations.
Firstly, policy design presented in this study seeks to engage with other questions. It will need to address the Big Why and the Big (for/with) Who, even more than the Big What. In strategic aspects, the What would need to be better aligned with the (by) Who, embrace open, social innovation and dialogic, strategic foresight. The strategic How cannot be “small” but it cannot be done separately from the other “big” questions. The “smallest” one — for governments — would need to be the operational Whats, Hows, and Whens. And most of it would increasingly be automated anyway, which further calls for government action on ensuring that this time it is not algorithms and Big Data (or the block chain technology) that would “creep-up” into strategy and policy.
Digital technology will continue to impact policy design, and even more. The world will not be taken over by robots, but humans will need to better do their human roles. Using digital to augment “humanness” and knowing why and how to effectively combine digital with analog, is a new horizon. Al will push us in that direction, but without a change in perspectives it might not end up being for the better. Already now there are warnings that we need to “take bias” out of algorithms, as if human would ever be able to do anything without a bias (or politics of it). It is not only due to the social reasoning and the specifically social nature of humanity, but also because policy and design and policy design are always based on value judgments, whether we are aware of it or not.
Nevertheless, we should also be aware of potential implications of political development on policy design. This does not relate only to the rapid deterioration of the three pillars of modern state (democratic accountability, rule of law, and public administration), but also the changing context of legitimacy. There used to be institutions that “preserve” or “conserve” public values, but those are being “innovated” on the model of the “Silicon Valley”, or further technocratized and managerialized. Moreover, the line between political campaigning and governing is further disappearing (the “continuous campaign”) and political parties (as the pillar of representative democracy) are becoming “government parties”. All of those trends would require more politics in policy design, and not less.
Secondly, there is certainly a need to go beyond government in considerations of policy design. Governments (or the state overall) have been “hollowed out” and even if bring it back “in”, governments would still be only one of the actors in broader governance. Most probably, we would experience much more of society-centric than state-centric governance in the future. Policy design ought to include other stakeholders — beyond “consultations” — as equal partners. They would still have different roles, but governments could hope, at best, to become a center of concentric circles (preferable a “liquid core”).
But beside the scope of public policy design, the methodological framework proposed here can be adjusted to design of policies in other institutions. Policy design might be used in public service delivery organizations (i.e. public sector[12]) beyond their role in public policy implementation through services. And there is no reason why communities or even corporations would not adopt it. They might need to go that direction anyway, with the recent shift of public preferences to the issues of values and social impact. For corporations, policy design might be a great opportunity for dealing with the “triple bottom line” or the “5Ps”. Practical opportunities are just around the corner, with regard to the recently adopted Agenda 2030 and the Sustainable Development Goals.
Most importantly, the role that political assemblies/parliaments played in history — before the rise of technocracy and the dominant role of governments in policy — might also need to be recovered. Policy design might eventually make genuine democratic Agoras (or Germanic “things” or Slavic “sabor”) out of modern legislatures and be the place for “drawing things together” for the betterment of society.
At some point, policy design might need to further open the doors go beyond “human design”. By recognizing Latour’s “collective” we might “bringing into” the design space the natural environment, and finally acknowledge the agency that objects have always had (i.e. the “vibrant matter”). There might even be a future in which the AI objects (in some other form, of course) become a sort of “subjects” that (who?) would be offered a “seat at the table” when “drawing things together” in policy design.
Discussion
Continued discussions on what policy, design, and policy design is or is not — while intellectually rewarding sometimes — cannot brings us closer to improving the way policies are designed. We should move from such scientific mindset and engage with actual design of what we collaboratively seek to bring forth into the world.
As Ackoff (2003) warns, improvements in the performance of all or some parts of the system taken separately may not — and often do not — improve the performance of the system overall. Often, such fragmented “point-solutions” may actually lead to further deterioration of the system. Along those lines, he also argues that we cannot know what we want merely by identifying what we do not want. This is not to mislead us into the “myth of comprehensiveness” (Nelson and Stolterman, 2012) — we cannot change the system as a whole, as we cannot even bring the “whole system into the room”. Thus, our shared work on policy design should be about seeking the “ultimate particular” in a constantly changing context.
An attempt to facilitate convergence of policy and culture is more promising if done in an oblique manner. Cultures change more organically with new practices, and design and policy cultures might not be as different as believed. Hence, we would need to be focusing on changing the ways of doing policy and doing design so as to help design and policy practices develop a shared approach converging into policy design.
Hence, this study represents only an intermediary step. Much more work is required to both articulate new understanding and make practical inroads to improving real-life applications. However, this should be done outside the conventional research and by actually designing new approaches in actual design situations. And that would necessitate committing to working on continuous innovation on the long haul. For that intention we need to bring together all the resources — human, technological, financial — and consolidate them into new policy design research initiative, enabled by platform frameworks.
Ultimately, it would require nurturing of a new kind of design[13] — policy design — and a new kind of designer — policy designer — in the context of a new governing context that might still prove to be a dramatic shift from what we have experienced in the past. Such investment would be worth making because we would develop capacities that would help produce multiplicative effect — rather than investing time and energy into seeking a “silver bullet”. Moreover, this would mean engaging with the “box” — rather than merely thinking “out of it” — and spending as little time as possible in “safe spaces”. Eventually, we might finally embrace Latour’s post-Promethean future and bring forth the humanity we desire.
It is now the time to do the second letting go; close; and consider the study satisficing for the moment. The end of this study is a new beginning that might start with a new design briefs on integrated policy aligning for the global 2030 Sustainable Development Agenda and exploration of opportunities and challenges that the platform approach might bring to policy design.
[1] And, additionally, the notion of value judgment in Özbekhan and the logic of political reasoning in Diesing
[2] Similar approaches can be found elsewhere, including in the assemblies of Slavic people (“zbor” or “sabor”).
[3] In the context of social complexity — and in line with the principles of “wicked” problems — framing of problems depends on anticipated solutions; identifying solutions is based on assumptions of the problem; and instruments (technology and methodology) influenced them both by intermediating between them.
[4] Refers to the system archetype: continuous decreasing of performance when settling for ever lower goals.
[5] See the original source Lindblom (1959 and then 1979), and then also Flach (2012) and Peters (2016).
[6] There is also the obstacle of separating the process from the content, as those are also interacting — see related considerations by Wittgenstein.
[7] Such “containment” is facilitated by two biases: limited participation based on limited perspective of what the problem represents; and limiting options based on problems being represented in terms of maximizing benefits and efficiency considerations.
[8] It would be probably too much to expect we could get to the 12th — where we reach “no paradigm” state of mind — but it might be that this time we will need a “system without a system” to realize our aspirations.
[9] From She Ji in his response to the proposal for the DesignX.
[10] From his lecture on “How to think like a state” at SciencesPo in 2007.
[11] More on this forthcoming, but essentially it implies working on long horizons by starting from the “next corner”, i.e. when the current cycle of trends will change into another one.
[12] I consider under public sector any organization — private or government owned/financed — that provides (instead of administrative) services to the public.
[13] It is important to note this is not reflected in academic curricula. OCAD’s MDes in SFI has a course in policy innovation (effectively, on administrative innovation) and there are several other design programs that lightly touch on policy. The Transition Design at the Carnegie Mellon provides some conceptual guidance for policy design issues. I am not aware of any program in policy studies that incorporates design. In Scandinavian countries, aspects of social design and of design for democracy mostly focus on community level.
