Keywords

The birth of modern public health is the flip-side of the coin of rapid urbanisation in the nineteenth century. Public health, urban governance and politics go together. More recently, the realisation that health is not created by the medical care delivery system and its associated industries has been rediscovered since Thomas McKeown (1976) showed that sewers, not drugs, improved population health. Public health adepts knew this since the German princely states’ Gesundheitspolizey (Gaißert, 1909), the occupational health equity analyses in France by Villermé (1840), Virchow’s aphorisms about politics being medicine writ large and of course British advances around removal of pump handles and early forms of Geographic Information Systems (Snow, 1855). Over the last half century, this call to cast the net wider, or even in different directions, has not relented.

Roughly since McKeown’s analysis, the health field has called for making health policy development the responsibility of all sectors, not just the health care systems. Such calls include the Alma Ata Declaration (World Health Organization [WHO], 1978), the Ottawa Charter for Health Promotion (WHO, Health Canada, and Canadian Public Health Association, 1986) and United Nations (UN) high-level ministerial statements on the control and management of chronic disease (e.g., by UN, WHO, cf. Glasgow & Schrecker, 2016). Interestingly, global governance parameters dictated that most of these statements and compacts aimed at the nations-state, with the exception of the Ottawa Charter—which embraced a ‘settings’ approach as it recognised that health is made not in the capital, but in the streets, corridors, schools and marketplaces of localities. The urgency to integrate echoes calls from administrative and political science, also first voiced in the 1970s, to join up policy systems. There is a range of monikers for either, from Healthy Public Policy and Health in All Policy to Whole-of-Government and Integrated Governance. Whatever it is called, it remains what Peters has called ‘the holy grail of public administration’. Wholesome integration is a good thing, and the WHO has enthusiastically been compiling vast series of case studies around successful intersectoral action (de Leeuw, 2021a).

But systematic appraisals of integrated health policy—and the processes that brought them about—are few and far between. The integration rhetoric seems strong, and the evidence appears light. The practical reality that a sewer line is not within the policy or operational remit of a medically qualified professional is immutable, but also of such an esoteric nature that its cognitive consequences are not successfully moved into other realms. The clinic does not run sewerage infrastructure. When Nancy Milio exhaustively documented how virtually every government sector impacted on health (1981a), the implication of a necessity to join up public policy for health was equally irrefutable. However, the practical demonstrations of such integrated policies were limited and deemed too unique to their contexts to be of replicable global relevance. The North Karelia project (Puska, 2002), the Norwegian Farm-Food-Nutrition approach (Milio, 1981b) and Heartbeat Wales (Nutbeam & Catford, 1987) were heralded as great successes of integrated approaches for health. But political science, administrative science and policy studies scholars have rarely applied a rigorous (theory-based) lens to explain these professed triumphs. Health promotion and public health policy appear to remain adrift on a sea of case studies.

Considering the place of ‘the local’ in the context of ‘the global’ is similarly challenging. Barber (2013) argues in a popular meme that cities collect the garbage, and that mayors should rule the world. The systematically compiled evidence that this makes sense, particularly in the health realm, continues to emerge.

FormalPara Chapter Navigation

The nature of this chapter is part scholarly, part personal journey. In my reflections on the use of political science in understanding social choice (i.e., policy) for health, my occasionally ill-informed career choices have led me to a place where I am convinced that, most of the time, local is better. Acuto and Leffel (2020) and Acuto et al. (2021), in fact share my views and add a strong conceptual call that local must also become global. This is my gaze: I hope to cast a political science look at local health processes—not with an ambition to be comprehensive but to illustrate that this yields superior insight. First, I illustrate the rise of local perspectives in health policymaking through a case study that describes agenda setting for health policy (also known as Healthy Public Policy, or—more recently—Health in All Policy). This research demonstrates that local health policy developments are delivering better processes and outcomes. From this, the argument progresses to look at policy analyses of the European WHO Healthy Cities network and contrasts some of its premises to other global urban health efforts. The key lesson seems to be that value-based (i.e., community inspired and supported, solidarity and equity driven, ecological and sustainability targeting) local health engagement seems better than a more traditional neo-liberal (i.e., new public management) Key Performance Indicator hinged approach. The question then becomes how successful local health policy in one spatial context may lead to policy learning and transfer elsewhere. The first stage in answering this question is found in a critique of mechanistic ‘knowledge translation approaches’. The second is a demonstration of the more switched-on political nature of political science approaches to policy learning. Here, I argue—based on our research in local health policy—that the very nature of local government and community allows for easier and more transparent policy inspiration. I make a case for policy development processes through mapping networks of policy language and policy actors and I argue that this would be easier at local level than elsewhere. Thirdly, there is a normative dimension. With over 15,000 ‘Healthy Cities’ (de Leeuw & Simos, 2017) and probably thousands more that are willingly struggling with the nature of health in their governance remits, there is simply a need to address the determinants of health (be they social, commercial, or political) with clearer guidance for political leadership at the local and global levels. Finally, with the reader joining my journey, you will also experience the increasingly complex nature of my theoretical choices and applications between 1986 and 2021… I have landed in a health political science space where an eclectic amalgamate of theoretical insights makes perfect sense. I am in good company, I feel: Paul Cairney (2013) suggests that there is added value to be found in added theories.

1 The Rise of Local

Based on pronouncements by Milio and Hancock, the Ottawa Charter for Health Promotion called for the ‘Building of Healthy Public Policy’ (Milio, 1981a). A natural sceptic, in 1986 I found this an intriguing message: would it really be possible for a nation-state to embrace the evidence—and then formulate and implement policy—that health and health equity can only be developed and boosted through truly integrated approaches? The Netherlands’ government had published its intentions in a discussion document—accompanied by an exhaustive series of sectoral background briefings—called ‘Nota 2000’. It took the Lalonde Report (Laframboise, 1990) one step further: all Ministries and public sectors were identified as having a role and responsibility in the promotion and maintenance of health. In 1989, my study of this Nota 2000 concluded that the development of ‘Healthy Public Policy’ at the level of the nation-state is virtually impossible (de Leeuw, 1989; de Leeuw & Polman, 1995). I arrived at this conclusion based on two core lines of reasoning.

First, I used Cobb and Elder’s agenda setting theory (Cobb & Elder, 1971, 1983). The two American political scientists, in the spirit of the day, had formulated a relatively straightforward and functional ‘strong theory’ (see Sabatier’s initial casting, 1999, for his views what such a theory ought to do). My inquiry concluded that ‘health policy’ (aka ‘Healthy Public Policy’) was not going to be endorsed and implemented at the national level in the country. Yet there was great enthusiasm for its potential at regional and local jurisdictional levels. In Table 1, the key propositions and predictive capabilities of the theory are presented (left column) as are some key inferences about how nation-state policy processes are qualitatively different from those lower jurisdictions (right column). I investigated the issue at the jurisdictional level of the nation-state. Surprisingly, and un-prompted, most interview respondents volunteered an opinion that Healthy Public Policy had more promise at the local government level (in the Netherlands, one of then ~700 municipalities) than nationally.

Table 1 Cobb & Elder agenda-setting parameters against their operations at local level
Full size table

Second, my analysis why the Netherlands’ national government was unable to develop, accept and implement a national Healthy Public Policy hinged on issues associated with power and, in particular, theories like the power-distance-reduction theory (Mulder, 1977; see also Harris et al., 2020). For Mulder, at the individual level:

  1. 1.

    More privileged individuals tend to try to preserve or to broaden their power distance from subordinates.

  2. 2.

    The larger their power distance is from a subordinate, the more the power holder would try to increase that distance.

  3. 3.

    Less powerful individuals try to decrease the power distance between themselves and their superiors.

  4. 4.

    The smaller the power distance, the more likely is the occurrence of less powerful individuals trying to reduce that distance.

When applied to institutional actors, the national level of policymaking was a far more stable (or rather, stale) environment than the local level where perceptions of power difference between the actors in the policy game were more easily overcome.Footnote 1 In terms of the punctuated equilibrium types of theories of the policy process (e.g., True et al., 2019), local policy processes seemed more punctuated and less balanced than national ones. At the local level, therefore, we could find faster responsiveness to (health) policy challenges. Local, in policy terms, is certainly quicker and may indeed be better.

Practical knowledge and procedural knowledge are at least as insightful as scholarly knowledge—and often more so. The ‘discovery’ that ‘health policy’ at the nation-state level was hard to accomplish was already foreshadowed by the visionaries behind the Ottawa Charter. In parallel to its development, they had accumulated a critical mass for a demonstration project at the local level (de Leeuw, 2017; Hancock, 2017). The World Health Organization in Europe, followed by (networks of) cities in North America, Australia and New Zealand, was exploding with enthusiasm for ‘Healthy Cities’, originally cast by Duhl (1963), formally posited by Hancock in 1984, substantiated by Hancock and Duhl in 1986, and boosted by Kickbusch and Tsouros into an urban social movement in the second half of the 1980s. These cities were to pursue eleven qualities (Table 2). Individually and as a network, they actively endeavoured to move from temporally and substantively limited projects into larger programmes and long-term policies for health (de Leeuw & Simos, 2017), even when their national governments failed to do so. The collateral by-catch of the national-level investigation revealed a much more exciting local opportunity.

Table 2 Hancock and Duhl (1986) evidence-based recommendations for the values of a healthy city
Full size table

De Leeuw et al. (2020) show that ‘Healthy Cities’ was the first transnational and global network of local governments and their communities pursuing a joint goal (before the mid-1980s international collaboration between local governments usually took the shape of more symbolic ‘twin city’ arrangements, e.g., Jayne et al., 2011, who also note that the context for city engagement in global affairs continues to change). Healthy Cities were followed by Sustainable Cities, a network formed in the lead-up to the Rio Earth Summit Conference (1992) and formalised in 1994 through commitments to an ‘Aalborg Charter’ and later a ‘Basque Declaration’ (cf Pinto et al., 2015). de Leeuw et al. (2020) analyse some of these ‘Theme City Networks’ against their stated impacts on health equity. They list in their review, among many others, Just Cities, Green Towns and Cities, Transition Towns and Ecodistricts, Winter Cities, Resilient Cities, Creative Cities, Knowledge Cities, Safe Cities and Communities, Festive Cities, Slow Cities as well as Happy Cities, Smart Cities, Child-friendly Cities, Age-friendly Cities, Conscious Cities and Inclusive Cities….

2 ‘Healthy Cities’ as Policy Code, and ‘Health Cities’ as a Rhetoric

In the—notably European—assessments of Healthy Cities, one thing becomes abundantly clear. Although ‘Healthy Cities’ adopt and build on an ambitious value system (including the pursuit of equity, solidarity, sustainability, empowerment, etc.), each locality follows its own path. These over 15,000 different paths are determined by history, culture, geopolitical connection, spatial dimension, growth and access to services, industrial and economic bases, etc. The maxim ‘If you’ve seen one Healthy City, you’ve seen one Healthy City’ is clear. In their essence, ‘Healthy Cities’ are localised health aspirations. The smallest self-declared ‘Healthy City’ is l’Isle-aux-Grues (a community on an island in the St. Lawrence River in Québec with around 200 inhabitants). The largest is the conurbation of Shanghai with over 16 million people. Brenner’s multi-scalar perspective on urbanity seems to be of particular relevance in understanding these thousands of diverse local health ambitions (e.g., Brenner, 2019).

In particular where there are strong and codified networks of Healthy Cities, there is coherence between the approaches and paradigms that they apply to their activities (de Leeuw, 2015). Beginning in 1986, the designated cities within the WHO/EURO Network had to formally commit to a clearly defined set of values, including the above Eleven Qualities. For each (approximately) 5-year ‘Phase’, European cities must commit to a collection of policy priorities set in connection with WHO’s global and regional (in this case European) work plans. For the current Seventh Phase of the European designated Healthy Cities network, these priorities are captured in the Six Ps (Fig. 1).

Fig. 1
figure 1

Current WHO/EURO Healthy City priorities (cf http://www.euro.who.int/en/health-topics/environment-and-health/urban-health/who-european-healthy-cities-network/healthy-cities-vision)

Full size image

Such a systematic and comprehensive approach to making Healthy Cities work is not universal for all the 15,000 self-identified Healthy Cities around the world. Currently, the Eastern Mediterranean Region of WHO is implementing a designation scheme focusing on the role of local government in Universal Health Coverage, and the Pan American Health Organization encourages local governments to achieve health equity through the systematic application of consultancy findings by Sir Michael Marmot (Rodríguez et al., 2019).

It is clear that different and relatively separate epistemic communities put their stamps on different ‘Healthy Cities’ networks (e.g., Goumans & Springett, 1997). This has in fact resulted in unproductive disconnects and often a focus on the development of policies and interventions that cannot necessarily be deemed the best or most appropriate ones. One of the most blatant disconnects is the one between European Healthy Cities (and its epistemic relatives around the world) and a Bloomberg/WHO Geneva sponsored network (see https://cities-spotlight.who.int/). The European Healthy Cities vision (including its global spin-offs in ‘la Francophonie’ including Africa, Oceania, Japan and South Korea, and Canada) hinges more on the above Eleven Qualities and is more obviously seen as a distal determinant of health, community driven, politically astute and value-based endeavour. The Bloomberg/WHO Healthy Cities Partnerships is connected to a market-oriented and quantitative perspective on the epidemiology of urban health and focuses on the prevention of disease (notably non-communicable disease) rather than processes of urbanisation and how they impact on the determinants of health. The language of the latter Healthy Cities approach supports this: they aim for ‘Best Buys’ rather than the former’s investment in sustainability, equity, community and solidarity. Kim et al. (2020) have determined that the label for the two ‘Healthy Cities’ is the same, but the paradigms driving their actions are diametrically opposed.

The vast diversity of these Healthy Cities creates an interesting dynamic and a vast potential resource for policy learning. An early evaluation of the European network of WHO Healthy Cities suggested that city networking made for more health policies that were more diverse, better implemented and took on board the value system associated with the social determinants of health, health justice and health equity (Camagni & Capello, 2004). Nation-state health policy diffusion has not been documented with rigour, as far as I know, although there is emergent research in tobacco control (e.g., Studlar, 2015). Locally, cities themselves claimed to have learned from each other and WHO how to integrate ‘upstream public health’ considerations in their policy repertoire (Farrington et al., 2015). The political science literature frames this phenomenon as policy learning or policy transfer. Hawkins et al. (2020) provide a concise overview of the pertinent literature. They frame effective policy learning and transfer as a multi-level governance challenge.

In the context of the premise of this chapter, it is important to make this point: policy learning and transfer (whether horizontal [from city to city] or vertical [from city to region, country and beyond]) are conditional on prevailing socio-cultural and political paradigms and associated epistemic communities. I will continue the exploration of the premise that local may be better by first looking at the nexus between research, policy and practice, and then consider a more socially dynamic understanding of policy transfer.

3 Translating Knowledge or Moving It Through the System?

In a more romantic world view—and many public health professionals espouse this as they believe that ‘health’ is an unchallenged aspiration of all humanity—there is an assumption that good evidence must automatically lead to good policy and its implementation. The underlying idea is borrowed from clinical research. It has led to what is commonly known as ‘implementation science’—not to be confused with policy implementation parameters… (Nilsen et al., 2013). New treatments and diagnostics need to diffuse and permeate into healthcare delivery systems. Individual and organisational behaviour change—often protocolised—are the tools of this trade. It uses language from the areas of diffusion of innovation and psychological behaviour change realms. Public health has similarly embraced this mantra of evidence-based policy and practice. But all too often solid evidence does not find an unobstructed way from research into practice, and practice is not adequately reflected in the scientific endeavour. This remains a frustration for the public health community which tends to operate, pragmatically, at the nexus between policy, research and practice. The gap between effectiveness on evidence, policy development, and practical intervention design and fidelity (implementing what was designed) has achieved increasing systematic attention in, for instance, Cochrane and Campbell Collaboration reviews.

The embrace of systematically generated evidence as a sine qua non for clinical practice has been powerful and pervasive. It has led to mechanistic clinical accountabilities, and some medical practitioners maintain that individual creative response to complex morbidities should transcend the results of Randomised Controlled Trials (Richter et al., 2020). Yet, the idea of ‘Knowledge Translation’ (KT) has become a major industry in the health field, essentially driven by a core logic of linear rational reasoning. Critics of the concept view it as a bad metaphor (Greenhalgh & Wieringa, 2011) that may have done the field more bad than good. ‘Translation’ as a metaphor would relate either to linguistics or to mathematics. Either one language (Clinician) is turned into another (Policy-ese), or in Euclidean geometry, a geometric transformation moves every point of a figure or a space by the same distance in a given direction, or as shifting the origin of the coordinate system. Neither of these is the perspective in health system implementation science, and its uncritical application to policy development may be considered fraught. The naïve view is also referred to as the ‘two communities’ hypothesis, an idea that has been rejected as mechanistic and stagnant (Lin & Gibson, 2003).

There are also conceptual and substantive problems with the KT suite of approaches (defined as a ‘dynamic and iterative process that includes synthesis, dissemination, exchange and ethically-sound application of knowledge to improve health’ (Straus et al., 2009, p. 165)). First, it is grounded in a presumed value-free Cartesian world view where facts are facts, and only facts matter. I and others have argued that facts, particularly in policy development and politics, are always subject to framing, morphing and negotiation. Facts are thoughts, thoughts are perceptions, perceptions are emotions, and we do not tend to think of emotions as facts. Cairney and Oliver (2020) build on this haiku and generate some evidence-based (but warranty-free) suggestions for engagement at the interface between scholarship, policy and practice.

I have called this interface the nexus. What happens at the nexus, and connects or separates the three domains of policy, research and practice can and should be studied. Understanding processes and structures that determine overlaps and gaps would enable us to generate better ways of generating knowledge for practice and policy. A systematic narrative review elicited two things about this arena (de Leeuw et al., 2007, 2008): (a) what tried-and-tested theoretical and conceptual models for work at the research-policy-practice nexus have been reported in the international peer-reviewed scholarly literature, and (b) are there organisations or groups that have a reputation for success in acting at the nexus, and do they follow the processes and parameters identified theoretically and conceptually?

Nearly thirty different theoretical frameworks specifically dealing with actions at the nexus were reported to have been applied. For analytical purposes, we grouped them into seven categories, which could then be put into three groups (Figs. 2 and 3).

Fig. 2
figure 2

Seven categories of theories and conceptual frameworks that explain what happens between research, policy and practice for health

Full size image
Fig. 3
figure 3

Graphical representation of seven categories of acting at the nexus between research, policy and practice

Full size image

The categories of practice-affirmed and tested conceptual models and theory-based evaluation build on and reinforce each other.

The Institutional Re-Design category of theories finds that to bridge the gap between research, policy and practice, you can set and enforce rules and other institutional arrangements. For instance, one could imagine that research is only financed once applied in practice (this would require a fundamentally different world view where base funding for research is guaranteed, and applied research rewardedFootnote 2) to secure immediacy and relevance. In the policy toolbox (with instruments in the areas of communicative, facilitative and regulatory intervention), we see a ‘Least Coercion Rule’ (Bemelmans-Videc et al., 2011)—policymakers tend to turn to rules that restrict behaviour only as a last resort. But Klijn and Koppenjan (2006), two policy network theorists, show that in particular network dynamics, the ‘rules’ can be changed. Actors engaged in policy networking may at times want to change the rules that formally or informally apply to the nature of, and the access to, the network, thus influencing policy outcomes. They may attempt to influence the shape of the network (by changing or consolidating actor relations, adding or changing procedures for access, or shifting external determinants of actor positions through, for instance, regulation), network outcomes (by changing performance indicators) and network interactions (by laying down instructions on conflict regulation or the governance of interaction). Hill and Hupe (2006) who see policy development and implementation mainly as a network governance challenge suggest that shifting rules is more easily achieved in local and other lower-level jurisdictions. This assertion resonates with the potency of street-level bureaucrats (see the updated Lipsky, 2010).

The Blurring the Boundaries model claims that it is possible to work towards evidence use in harmonious rather than conflictual ways, through trust, understanding and confidence between researchers, along with enhancing opportunities for research uptake. This model rejects the idea that there is a separation between scholars, practitioners and policy developers. Ideally, understanding ‘the other’ facilitates the development of shared understandings between these communities. By deliberately obfuscating organisational accountabilities and governance parameters this model would allow for true co-owning of research and policy processes (e.g., van Buuren & Edelenbos, 2004). Nahapiet and Ghoshal (1998) show that deliberate blurring of organisational and conceptual boundaries in the long run creates joint language and vocabulary that facilitates easier joint policy development and implementation.

The Utilitarian Evidence model states that only research products that are seen to be useful will be applied in policy and practice. This model describes how principles for the utility of research are different between researchers, practitioners and policymakers. It is important to recognise that utility is a dynamically perceptual quality and that its framing is as important as the ‘factual’ usefulness (see de Leeuw et al., 2018).

Fourth is the Conduit model. The ‘conduit’ informs different communities—policy communities, practice communities, the ‘general’ community—of research developments and outcomes. The conduit can be a person, agency or structure. A ‘conduit’ works to disseminate new knowledge in a format that is accessible and acceptable across groups (e.g., using more common, every-day terms, using graphs, avoiding jargon). The ‘conduit’ agent facilitates collaboration between the communities for the ongoing engagement of all partners in research (Bernier et al., 2006). The ‘conduit’ is an advocate and provides a platform for communities to express their concerns, in particular those who have fewer material and symbolic (e.g., skills and resilience) resources. Also, in disseminating new knowledge in an accessible manner, ‘conduits’ are at the ready to feed knowledge into fertile ground.

Sometimes research outcomes are not at all consistent with current political agendas or organisational practice. Alternative Evidence says that if research findings run counter to current political agendas/paradigms, its immediate potential impact will be muted. However, there may come a time where the volume of counter evidence can no longer be ignored—or at least not without creating organisational and political upsets or outrage (Hanney et al., 2003). In any event, researchers should also keep in mind that ‘at the end of the day, policies…are constantly framed and reframed in response to changing contexts’ (Choi et al., 2005). This model suggests that scholars and policy entrepreneurs should arm themselves with a repertoire/arsenal of evidence that can be inserted into the policy process when the opportunity arises.

Research Narratives aim to create a human dimension to research by including personal stories. Through personal stories, they inject ‘common man’ experiences into research outcomes (Sutton, 1999). The narratives humanise the research, but can also bring a sense of immediacy to the research topic that a ‘dry’ presentation of results might otherwise lack. Given policymakers’ wish to include experience and common sense (over esoteric science) in their ‘selection’ of evidence (Booth, 1988), the inclusion of narratives in the overall presentation of research would be appropriate. The narratives support the research, and they highlight practitioner experiences. Research Narratives approaches provide an additional layer to the previous four models.

The Resonance model works on the idea that researchers, policy entrepreneurs or evidence conduits should have their ‘finger on the pulse’ of belief systems. In doing so, they can link their research outcomes with popular or emergent belief systems (e.g., ‘social inclusion’, a ‘safe environment for all individuals’). When research resonates with what people believe, they find it easier to accept evidence.

Discourses around ‘morally fraught’ issues such as HIV/AIDS, birth control or euthanasia have often been framed from a religious starting point. It would not be helpful to argue that moral foundations are ‘wrong’ (and thereby polarise the policy discourse), as they are strongly connected to people’s life worlds. However, trying to make the evidence resonate with other belief systems could advance the application of new knowledge. The Research Resonance model argues, for instance, that connecting the HIV/AIDS discourse to issues of ‘safety’, and the euthanasia discourse to ‘dignity’, rather than to ‘morality’, is helpful in integrating research, policy and practice. Issues of safety and dignity are issues that any individual, irrespective of their belief system, can identify with. The Research Resonance model demonstrates how the ‘spin’ which promotes research can influence the level of public and organisational interest in the research.

Is local better in this particular view of ‘acting at the nexus’ (rather than ‘knowledge translation’)? Our suite of studies suggests it is—even at the institutional redesign level. The other six models are exquisitely well-tuned to the particular context of street-level engagement, short lines of accountability and communication between research agents and policy actors, and practical blurred boundary spanning. But even at the institutional redesign level (which most would conceptualise as a state effort), we see that, through Hill and Hupe’s (2006) gaze of multi-level governance, there is prominence and legitimacy for local types of governance (see, e.g., de Leeuw, 2015).

4 Policy Transfer: Scaling up and Scaling Wide

Earlier I illustrated policy agenda setting with work that applied Cobb and Elder (1971). A different perspective on how policies come about is provided by the policy transfer perspective. The suite of policy transfer theories and conceptual frameworks continues to be much debated and refined. The first comprehensive theory was proposed by Dolowitz and Marsh (1996). They argued that policy transfer occurs as ‘a result of strategic decisions taken by actors inside and outside of government’ (1996, p. 343). They saw that diffusion theory, policy learning and adaptation were all part of a bigger policy transfer process that describes the travel of policy ambitions through larger systems. Their most widely cited definition of policy transfer is:

… a process in which knowledge about policies, administrative arrangements, institutions etc. in one time and/or place is used in the development of policies, administrative arrangements and institutions in another time and/or place. (Dolowitz & Marsh, 1996, p. 344)

Such policy transfer may happen ‘voluntarily’ or ‘coercively’ (terms applied by Dolowitz and Marsh—for local policy we might think of a slightly different casting, e.g., ‘internally’ vs ‘externally motivated’). The coercive model happens where a government or supra-local agency ‘forces’ another government to adopt a particular policy that is in their interests, explicitly highlighting the importance of agency (or lack thereof) in their framework.

Making the distinction between voluntary and coercive transfer the core issue, Dolowitz and Marsh (1996) argue that voluntary transfer is the policymakers’ motivational result of some form of dissatisfaction or problem with the status quo. Perceived policy ‘failure’ creates incentives to identify new policies that can be borrowed from elsewhere—and at the local level, policy failure (a burst sewerage line; persistent health inequities among particular groups; etc.) is always more urgent and visible. Coercive transfer, while rarer, comes about through supranational mechanisms such as international treaties, trade and investment agreements, or the actions of international organisations such as the World Bank or International Monetary Fund (IMF). On occasion, the enforcement is carried by civil society through lawsuits, for instance in the Netherlands. Here, climate change advocacy group Urgenda successfully sued the state to act on its commitment to international agreements (Mayer, 2019). Law scholars see this case as deeply influential and suggest it will eventually impact on every level of government and force governance adherence to global standards—which then become local norms. This also works the other way around. The moment former President Trump withdrew the United States from the climate change accords, hundreds of state and local governments signed up to them—often with much more significant ambitions (e.g., Murthy, 2019).

Stone et al. (2020) suggest that policy substance can be instrumental in the transfer dynamic, and not just institutional politicking. An example of this dimension of policy learning and transfer can be identified in European Healthy Cities. The strict designation and accreditation processes are factors in ‘externally motivated’ policy development (de Leeuw & Skovgaard, 2005). Our analysis showed that the imposition of a designation process itself significantly strengthens adherence to local health policy objectives and processes. Interestingly, WHO maintains its position as an international collaborative member organisation and at best might facilitate coercion—except in cases where strong Treaty powers have been established (such as, for instance, the International Health Regulations, and less coercively, the Framework Convention on Tobacco Control [FCTC]). Indirect coercive transfer can be forced upon governments through other externalities, such as environmental damage, or through technological progress or economic integration, which compel governments to work together to solve supranational problems. How these commitments trickle down to local policymaking is as yet doubtful, but the increasing presence of networks of local governments in formal international forums suggests that we are on the brink of a glocal governance shift (see also Acuto & Leffel, 2020).

The literature identifies nine categories of transfer actors, some directly involved in the transfer process, and some who are external ‘influencers’: elected officials; political parties; bureaucrats/civil servants; pressure groups; policy entrepreneurs and experts; transnational corporations; think tanks; and supranational governmental and non-governmental institutions and consultants. This final group seems to have gained a particular power in recent years. Network governance and management concepts now show that policymaking is ever more becoming a non-state actor (including civil society and industry entities) enterprise. Provan and Kenis (2008) show the organisational and accountability mechanisms associated with this reality—and argue that lower-level government and organisational-level engagement are more elegantly suited to get this right.

Back to ‘policy transfer’—what precisely is transferred: the policy; its intervention package; an idea; institutional arrangements; or ambitions? Answering this question depends on our view of the nature of policy and the policy process. Dolowitz and Marsh (1996) seek to move away from narrow conceptualisations of policy and policy transfer and seek to incorporate broader macro-objects of transfer such as ideology, ideas and negative lessons. This seems to be a pertinent point to our Healthy Cities perspectives.

‘Transfer’ assumes a source and a destination. Although the destination may well be clearly identified (e.g., a particular level of government with clear governance arrangements), the source may be more diffuse. It is argued that policy actors can turn to different levels of governance—the international, national and local—to draw inspiration, which provides opportunities for both horizontal transfer (between nations, geographical areas or sectors) and vertical transfer (through different levels of governance).

This more diffuse image of the policy transfer endeavour also offers a spectrum of policy transfer modalities, from ‘carbon copying’ at one end of the scale to ‘inspiration’ at the other, where policymakers do not adopt all aspects of a policy or seek to achieve identical outcomes.

Dolowitz and Marsh (1996) emphasise that policies are not imported into a vacuum, and the contextual, institutional and political arrangements of the borrowing jurisdiction and the specific motivations and objectives of decision makers influence the success or failure of introducing policy and its implementation. Closely related to this issue, three elements of the transfer process have been identified that are critical for ‘successful’ transfer outcomes: information deficit, ‘cherry-picking’ and contextual factors. If a borrowing jurisdiction does not have full information about the policy itself and the institutional arrangement within which it sits, this leads to ‘uninformed’ transfers. Where a policy and the institutional aspects that make it a success are not transferred in their entirety, then there is the risk of ‘incomplete’ transfer. Finally, the differences between the original and borrowing jurisdiction can result in ‘inappropriate’ transfer. From our earlier research in Healthy Cities, it appears that the mere requirement of European cities to be active members of a national and the international WHO network was influential for its success and survival (Camagni & Capello, 2004): the more they networked, the better they seemed able to deliver on Healthy City qualities.

The Dolowitz and Marsh model faced criticism of its status as a mere heuristic device rather than an explanatory theory of policy change (Evans & Davies, 1999; James & Lodge, 2003). Evans and Davies (1999) elevated these ideas and merged them into a perspective that embraced a connected and nested approach to policy movement through multi-scalar systems. They see five dimensions:

  1. 1.

    International structure and agency;

  2. 2.

    Domestic structure and agency;

  3. 3.

    Policy network analysis;

  4. 4.

    Policy transfer analysis; and

  5. 5.

    Epistemic community approaches.

Of these, in Healthy Cities evaluation work that I have led and carried out, we adopted (1), (3) and (5). Evans and Davies also focus on the ‘spatial’ dimension of policy transfer within and between different levels of governance. They identify 25 transfer pathways, working horizontally and vertically between the five different levels of governance (the transnational, regional, international, national and local levels). They argue that economic, technological, ideological and institutional structures of the ‘borrowing’ and ‘lending’ jurisdictions must be analysed in order to establish how they facilitate policy transfer and the impact on the transfer process.

They end up with a model of twelve stages. In these, the transfer process is broken up into discrete (but connected) actions:

  • Recognition: whereby a decision-making elite, politician or bureaucrat (known as the ‘client’) identifies a policy problem;

  • Search: which is undertaken if an obvious acceptable policy response is not available;

  • Contact: whereby a policy transfer ‘agent’ (e.g., an epistemic community within an international organisation) is identified;

  • Emergence of an information feeder network: whereby the ‘client’ is provided with an increase in volume and detail of information regarding the potential for transfer;

  • Cognition, reception and the emergence of a transfer network: whereby the client evaluates the information provided by the information feeder network, with cognition and reception depending on a common value system existing between the client and the network;

  • Elite and cognitive mobilisation: whereby the transfer agent’s information and networks are tested, as they are expected to provide robust information on programmes or policies that address similar problems to those experienced by the client at the ‘recognition’ stage;

  • Interaction: whereby the transfer agent organises forums for the exchange of ideas between the client and knowledge elites with policy-relevant knowledge;

  • Evaluation: whereby the client evaluates the intelligence gathered by the agent relating to the object of transfer (e.g., policy goals, content, instruments, institutions, ideology), the degree of transfer (e.g., copying, hybridisation or inspiration) and the prerequisites of transfer (e.g., political feasibility and institutional conditions);

  • Decision: whereby the chosen policy is tested against other competing ideas arising in the borrowing jurisdiction, within what Kingdon (1984) describes as the ‘policy primeval soup’;

  • Implementation: whereby the policy is adopted by the client country, often by implementers who are different people to those who formulated the policy.

In the peer-reviewed and more narrative accounts of Healthy Cities success (e.g., Farrington et al., 2015; Tsouros, 1991), these action areas are clearly identifiable. Interestingly, they may not have resulted in a longer lasting or even permanent institutionalisation of local movements for health—they are morphing and shifting, as always. de Leeuw et al. (2020) in fact describe how the health generating potential of other glocal networks (notably Citta Slow and Sustainable Cities) may be more significant than Healthy Cities a l’Europe.

5 Framing and Entrepreneurship for Network and Systems Change

This, then, brings us to the next political science-inspired question: how do Evans and Davies’ (1999) twelve actions create enduring and meaningful policy change for local health? In a theoretical and methodological ‘proof-of-concept’ paper (de Leeuw et al., 2018), we attempted to demonstrate and map some of the cognitive-informational and network dimensions of their approach. We wrote that actor networks and frame networks could (or should) be interleaved to identify opportunities for boundary spanning and policy transfer entrepreneurship.

The policy network map or configuration does not necessarily indicate which dynamics of the policy process will create new opportunities for policy development or voluntary transfer. Evidence-informed (or based) policy change does not just depend on network structure, but also on actions to shape the dimensions of the network. That is, the processes of negotiation and cooperation which result in the identification and pursuit of new ambitions to resolve social issues must also be examined (cf Leftwich, 1994). In other words: what do individuals and organisations do to change the network configuration to their (policy preference) advantage?

This is not the place to systematically review the body of literature around policy change actions, and I will necessarily remain brief and eclectic. Laumann and Knoke (1987) mapped two policy domains in the United States (health and energy). They found that actors that deploy more personnel to scan and anticipate organisational behaviour and particular policy process interventions of the other network elements are better able to realise their ambitions—they anticipate, pre-empt and counter policy process change. John Kingdon (1984) famously identified ‘policy entrepreneurs’ who engage in processes of ‘alternative specification’ to connect actors and events in the eponymous Multiple Streams. Those individuals, who well may be functions of social and political entrepreneurial institutions, are variously described (e.g., Skok, 1995) as ‘issue initiator’, ‘policy broker’, ‘strategist’ or ‘caretaker’ in addition to ‘boundary worker’ and ‘issue manager’—they connect, disconnect and reconnect the players in the network around particular versions of the same reality (Knight & Lyall, 2013). The local health policy research that I cover in this chapter (e.g., also de Leeuw & Lin, 2017; Hoeijmakers et al., 2007) shows that these perspectives and roles are both more astute and tangible at the local level than they can be at the national or global level.

Common to this dimension of policy agency is the capacity to deploy relevant language. This, again, is not a new theoretical proposition in the evidence-policy-practice discourse. Stone (1997) dwells extensively on the power of the word, rhetoric and symbolism in policy processes; Khayatzadeh-Mahani et al. (2017) recently showed that words can act as ‘collaboration magnets’, and the novel behavioural economics units springing up around the world to nudge the citizenry to adopt preferred government action (Frain & Tame, 2017) use language—rather than monetary incentive or other ‘hard’ policy instruments—as their most prominent change tool. In fact, the particular framing of (perceived) reality is a key technique for individuals, groups, communities and political systems to make sense of the world (Entman, 1993). The art of mastering discourse, framing and storytelling may well be key to successfully bridging the nexus between whatever is construed as ‘the evidence’ and the negotiated endeavour to resolve social problems (i.e., ‘policy’) as Davidson (2017) astutely demonstrates. He quotes a tweet by Advocate Thuli Madonsela, former Public Protector of South Africa (4 July 2017):

“For people who want the truth adequate evidence is enough but for those who don’t want the truth overwhelming evidence is inadequate” which echoes an almost proverbial insight by famed economist John Maynard Keynes: “There is nothing a Government hates more than to be well-informed; for it makes the process of arriving at decisions much more complicated and difficult”. (Keynes & Moggridge, 1982)

To use an ancient Greek perspective on ‘knowing’, it may just be episteme ‘the facts’, but also regna ‘wisdom’, phronesis ‘political astuteness’, techne ‘skill’ or even parrhesia ‘speaking truth to power’ and combinations thereof (Sharpe, 2007) that create the urge for policy change. It appears that effective operators, notably at the local level, recognise two things:

  • We are dealing with complex, dynamic, interdependent and interrelated groups of actors driving or affected by the policy process;

  • Among the many tools in the process of shaping policy processes, the understanding of the discourse and the mobilisation and resonance of language and symbols are key (Pearce et al., 2014).

Drawing on a symbolic interactionist perspective, I posit that cliques or clusters identifiable through policy network analyses also share world views constructed in language, symbols and frames. For instance, in the policy network that would describe public choice around particular diagnostics in internal medicine, the clusters might comprise radiographers, gastroenterologists, hospital administrators, laboratory personnel and health economists. They each have their views of what creates ‘evidence’. In the context of the discussion above of five dimensions of policy transfer, theorists like Haas (1992) would call them an epistemic community; Laumann and Knoke (1987) would simply frame them as a policy sub-domain. Each group has their particular training, disciplinary grounding, professional affiliations and accreditation mechanisms, and views of what constitutes ‘truth’. The feasibility of a policy development process to yield tangible policy outcomes would be greatly enhanced if significant chunks of language/symbols/frames are shared across the structure of the network.

In our proof-of-concept work, we showed that policy network structure and agency can be separated and then interleaved, to show how diverse positions in policy network elites can connect for policy change. In the nexus categories, I described earlier that we encountered boundary spanners and policy entrepreneurs. Effective ones can see both the structure (network configuration) and engage with the discourse (the set of frames pertinent to the policy network domain) of a policy development environment: if the unique language representations of each clique would not share any commonality, there would be no reason or opportunity to meaningfully engage in any policy discourse. In short, we would be interested in both the boundaries and the spanning parameters of what only seems to be a messy enterprise (Rütten et al., 2017).

The extant literature on boundary spanning tends to focus on complex issues of public service provision, in the words of Williams (2011, p. 27) involving:

“people and organizations working together to manage and tackle common issues, to promote better co-ordination and integration of public services, to reduce duplication, to make the best use of scarce resources and to meet gaps in service provision and to satisfy unmet needs.” Boundary spanners – in service provision or policy engagement – occupy interconnected roles of reticulist, entrepreneur, interpreter/communicator and organizer. (Williams, 2002)

Our proof-of-concept analysis, based on a single health policy case study, demonstrated the feasibility of integrating the analysis of network structure and one form of network agency (Browne et al., 2017). An important area for future research is the role of the boundary spanners/policy entrepreneurs/policy brokers, identified using these network approaches. Based on our analysis, we hypothesised that the points at which frames overlap represent opportunities for boundary spanners to shift the policy discourse and, in turn, reconfigure the network. Now that we have demonstrated that dual analysis of the structure and agency of policy networks is feasible, the potential for boundary spanners to bridge the policy-evidence nexus needs to be tested empirically. Such research would be valuable for advancing policy network theory and may assist grassroots organisations engaging in advocacy to more effectively influence the policy agenda and to integrate evidence into policymaking.

The terrain of moving ‘evidence’ into ‘policy’ is not a one-dimensional map that cannot be navigated with traditional scientific insight alone, as the core of the Knowledge Translation perspective claims. Its landscape is multidimensional and can only be fully appreciated when its ever-changing nature is taken into account. Our efforts of overlaying the network structures and agencies of the policy narrative topography add a potentially important compass to the toolkit of the boundary spanner and policy entrepreneur.

6 How is Local Better?

Local is better. It is not just where the garbage is collected (Barber, 2013). Policy failure and success are more easily identified and communicated. The diversity of stakeholders is potentially more easily identifiable and to be engaged. The nexus between research, policy and practice becomes visible and potentially more acute. Problems need solving, and policies can be made. The organisation of network governance and management is more straightforward, and policy transfer can happen between stops on a bus route or metro line as jurisdictions are closer and tighter. Nations-states are beholden to rigid welfare state paradigms, and localities have the potential to break free (de Leeuw, 2021b).

Of course, this is a rosy casting of the issue. Many challenges remain, with local institutions failing to be transparent and/or retreating into hermetic bureaucratic walled compounds. Society itself may also fail to embrace and exploit its opportunities. With the ever-reducing degrees of social capital (Putnam, 2000), a stifling indolence has descended on communities around the world, where inward-looking egotism has replaced a sense and belief in community. Community engagement in policymaking, participation in democratic institutions and having one’s voice heard for systems change seem to be more remote ideals than ever. The internet age has driven many into the arms of clicktivism and hacktivism as most individual expressions of social concern (George & Leidner, 2019). The research dimension of the nexus has come under pressure of framings on alternative facts and fake news (de Leeuw, 2018). And yet, local is better.

Local government and local institutions (like sports clubs, dog parks, cafes and restaurants, hairdressers) are still the flashpoints of political organisation and the exchange of social and political emotion. There is a role for everyone at those physical and virtual venues. They create and sustain health—very much in the spirit of the Ottawa Charter’s ‘settings’ gaze. Good local government—and good local governance—still has an opportunity to challenge, build and exploit the potential of civil society. This even extends beyond what is regularly labelled as ‘the citizenry’: we tend to forget (and in fact have assumed as implicit in our preceding argument) that ‘locals’ are ‘citizens’. But in many parts of the world the locals do not have rights and very limited visibility. They live, as slum dwellers, in hidden cities (WHO & UN Habitat, 2010). But for them, without local government support, technology comes to the rescue through IT-enhanced mobile phone applications that generate data and create voice (Corburn & Karanja, 2014).

Local is better—because of sharper policy agenda setting, opportunities of making power differentials more visible, exposure to policy learning and transfer, and the mere fact that everyone has a better opportunity to be heard. We can engage in better policies for better local health through mutual respect in open governance networks and through the systematic development and application of theoretical and empirical frameworks from political science.