Is ‘Policy Diffusion’ Genuinely an Explanation or is it a Metaphor for Policy Change?

In an ever-increasing interdependent world, the number of policies bearing resemblance to those located within different jurisdictions is highly prevalent. Such convergence explanations that offer a more or less accidental accumulation of independent national reactions to globalisation, modernising forces and contemporary phenomenon fall short of a through examination into how this has taken place. It fails to recognise the density of communicative interlinks and symbiotic relationships that exist in the modern world which make isolated national policy decisions increasingly unlikely. Against this background, theories of diffusion have emerged as a challenge to narrow explanations and as a distinct class of convergence mechanisms to be systematically integrated in the analyses of policy convergence. Despite its many proponents, such policy transfer literature has also been criticised for being “overtheorized and underapplied” (Bennet and Howlett, 1992). This paper aims to interrogate what policy diffusion entails and examine its application. Following an overview of four contending forms diffusion can take; coercion, competition, learning and emulation, this paper will question the limitations of relying on concepts of diffusion for policy change.

What is Policy Diffusion?

From the beginning of history, analysts have examined how other nations deal with problems that their own nations have suffered. Alexis de Tocqueville’s detailed analysis in Democracy in America, first published in 1835, could be seen as spurred by concerns within his home country, France, given its ‘difficult transition from aristocracy to democracy’ (Browers, 2003). Such a comparative analysis can lay out alternative policy approaches for policymakers, which may not have first been considered domestically. Diffusion can be defined as a process by which policy innovations are communicated and adopted voluntarily over time. Diffusion processes have received considerable attention in explaining policy convergence at the national level (Walker 1969; Gray 1973; Berry and Berry 1999). However very few issues occur in one country alone and policy diffusion, in its recent studied form, looks to the international spread of policy innovations. This contrasts with collective decision-making within international institutions and as such, the essential feature of policy diffusion is that it occurs in the absence of formal or contractual obligation. It is a horizontal process whereby individually adopted policies and programmes add up to a decentralized regulatory structure (Levi-Faur 2005). Unlike in the case of multilateral legal treaties, which are negotiated centrally between states and subsequently implemented top-down, with diffusion, decision-making procedures are decentralized and remain at the national level.

Certain understandings of diffusion view it as “any pattern of successive adoption of a policy innovation”, regardless of the casual mechanisms through which these adaptions were brought about (Eyestone 1977). This is similar to its definition in the natural sciences, where molecules spread from an area of high concentration to an area of low concentration. However, this interpretation’s analytical usefulness for the study of policy change and convergence is rather limited. Instead of explaining how policies spread internationally, it merely provides a new linguistic term for the rather trivial observation that policies actually spread from one country to another. For this reason, it is necessary to instead examine the mechanisms and processes that are associated with the likelihood of policy diffusion. I will therefore examine four contending methods of policy diffusion that are supported by a number of scholarly efforts; Coercion, Competition, Learning and Emulation (Braun and Gilardi, 2006; Simmons et al., 2006; Shipan and Volden, 2008; Gilardi, 2012).

Coercion

Coercive diffusion occurs following the adoption of a policy under pressurising or threatening conditions to do so. This could include financial reasons, such as the promise funds or the removal of others, the application of legal actions or even the use of military interventions (Rose, 2005). Historically, the imposition of policies during colonial eras is an obvious example of coercive policy adoption at work. Contemporary, we can see elements of it in situations where democratic nations are compelled to follow the rules of larger institutions to which they are members including the EU, international human rights organisations or the World Trade Organisation. Actors involved in coercion differ fundamentally in their motivations, which generally lie in exporting their fundamental values and principles, but also in the extent to which they can exert such pressures and shape adopted policies. In the majority of instances, policymakers in the receiving country have little space to influence the policy content and as such they follow what has been determined by the imposing actors, sometimes due to expected political or economic benefits. Despite its hold, strong governments do not always necessarily succumb to pressures and may refuse to implement certain policies. This was witnessed with France and Germany’s evasion of the stipulations of the EU’s Stability and Growth Pact in 2003. The pact’s credibility suffered as the two biggest euroland economies blocked moves to impose fines for exceeding budget deficit limits, this was despite weaker states such as Spain undergoing a coercive process and complying (Tran, 2003). Certain scholars have argued that coercion should not be considered a process of diffusion as it ignores a key element of interdependence, the enthusiastic willingness for a decisionmaker for policy adoption in reacting country (Wasserfallen, 2017).

Competition

It must be remembered that the policy learning process is not an apolitical one, and that what “is “learned” and what is “remembered” must always be seen in the context of political interests and political power” (Bennett and Howlett, 1992). Economic competition can lead to the diffusion of policies when there are economic spill overs across jurisdictions. In this case, the changes are wrought not by powerful actors, but by direct competitors who compete for capital and export markets. It is a given that governments must choose market-friendly policies if they wish to attract global investment and retain a competitive export industry. This is amplified when a nations direct competitor has also done so, for example, if a country’s competitors simplify their regulatory requirements or perhaps reduce tax burdens. Research by Swank (2006) and Genschel (2002) show diffusion of nominally lower corporate tax rates among OECD countries after the United States reduced corporate rates in the early 1980s. Initial local political resistance was influenced when countries jumped on the bandwagon. Another policy area which has been examined in terms of the effect of competitive diffusion is environmental regulation. An almost ‘race-to-the-bottom’ view has emerged to explain countries lack of will for enforcing environmental protectionism on the operations of firms in their jurisdictions. The expense of complying has allowed the possibility for firms to threaten to relocate and dump production activities in developing countries which offer lax regulation (Porter, 1999; Tanguay, 2001; Wheeler, 2001). This can be linked to forms of coercive diffusion but explicitly relates to economic reasons and competitions for capital.

Learning

Hall defines the ‘social learning’ as a ‘deliberate attempt to adjust the goals or techniques of policy in the light of the consequences of past policy and new information to better attain the ultimate objects of governance’ (Hall, 1993). The process of policy diffusion through learning assumes that policymakers follow this logic. For Rose (2005), learning enables policymakers to move beyond trial-and-error based learning developed from their own experience and conjectures about what ‘might’ happen, towards the direct observation of programmes already in practice in other settings. In the form of ‘lesson-drawing’, policymakers create a model of the policy they wish to draw lessons from and programmes in effect elsewhere are scanned and stipulated on what would happen if they were to be transferred to another country. With the peace talks in Northern Ireland leading to the Good Friday Agreement of April 1998, Irish Tánaiste Dick Spring examined the use of ‘proximity talks’ in the Dayton peace negotiations over Bosnia. It involved the use of multiparty talks without a ‘face to face’ element. Policymakers in Northern Ireland also studied attempts to come to terms with a violent past in Latin and Central America and South Africa (Darby and MacGinty, 2001). In the case of best practice learning, one specific policy approach is identified, operating in one particular nation, and promoted to other nations as the most effective. This is a rather prescriptive approach to policy learning and holds the assumption that a specific policy can be awarded the title of ‘most effective’ or the ‘best’. This contrasts with another form learning can take, benchmarking. Benchmarking enables the comparison of different policy approaches with each other, without necessarily normatively labelling any one approach as superior to all others.

Emulation

Inspired by sociological research and constructivism, emulation often occurs when few guides to a policy development already exist and where outcome conditions are highly uncertain. DiMaggio and Powell (1983) highlight this in their notion of “mimetic isomorphism” in which an organisation will imitate another organisation’s structure which is perceived as legitimate as a guaranteed ‘safe’ way to proceed. Norms and conventions are socially constructed, and policy makers conform to these with the adoption of appropriate policies (Finnemore and Sikkink, 2001). It is often the low-cost approach to policy making, mimicking what may happen due to common characteristics socially, culturally or geographic proximity. However, important to note is that emulation is not necessarily the result of replicating the best practices of other nations such as the case of best practise learning. One policy area which has fallen into this category many times is the environment, with policies being “fitted fairly randomly into political problems when the opportunities are favourable” (Jordan et al., 2003). Zhou et al. (2019) in their study of renewable energy policies in Europe and Bisenbender and Tosun (2014) in their examination of air emission standards both found that adoption is positively driven by normative emulation from neighbouring countries. Emulation involves actors seeking policy implementation which can be regarded as ‘good enough’ rather than striving to obtain policies which can be considered fully optimal. In administration theory, these insights are offered by Herbert Simon who argued that it is “the theory of intended and bounded rationality – of the behaviour of human beings who satisfice because they have not the wits to maximise” (Simon, 1957). Such bounded rationality employed by policymakers rest on psychological factors by policymakers, including an ill understanding of total options and their consequences and organisational factors such as structures which frame how policies are made.

Limitations of Diffusion:

Analytical Problems

Testing Convergence

To the majority of critics of diffusion studies, issues with analytically examining its existence is regarded as its biggest shortfall. Issue is regularly taken with how researchers analytically test convergence through diffusion. Such studies have examined single and comparative case studies based on personal observations of convergence, rather than testing beliefs. Plumper and Schneider (2009) overview articles which study policy convergence and found that 13 out of 27 did not employ analytical methods. One such example is a paper carried out by Dolowitz (1997) examining public policy, political development and democratisation. In the area of public policy, he notes that “much of the UK Conservatives’ welfare-to-work program was transferred from the US.” With political development he draws attention to the fact “as a condition for approving a loan, the International Monetary Fund often requires that a recipient country pursue particular economic policies.” Finally, in examining democratisation he acknowledges the case of “the Spanish constitution, particularly the sections dealing with the roles of the prime minister, the legislature and the president was modelled on the German constitution.” However, we must be aware that these examples given do not necessarily prove that diffusion is a leading explanatory factor in each of these issue areas. There is a lack of a probabilistic sampling technique to choose cases at random. Instead, they all involve positive instances where transfer has occurred without convincing evidence that a complete interrogation has taken place. Examining diffusion as an explanatory factor for many policies in a specific sector or during a specific timeframe would be helpful. Similarly, in cases of diffusion taking place in the form of economic competition between states, it is found that research tends to examine the openness of a country’s markets rather than the actual pressure that is exerted by competitors. Dobbin et al. (2007) suggests “empirical tests typically show evidence of diffusion of the latest policy fad, without providing evidence that policy makers had hard evidence that the policy in question provided the purported benefit for previous adopters.” In such a case a more accurate measure would involve testing “whether a country’s actual competitors for a specific good (e.g., foreign investment) have adopted the policy in question, potentially increasing competitive pressure.”

Differing Approaches

Once case studies are selected, analytical problems can also overlap with further problems of how scholars approach their research. Plumper and Schneider (2009) discuss the reliance on using a variance approach in interpreting convergence. Under variance, convergence is noted as a decline in the variance of observations. However, this can only work where “the process is unconditional, and researchers know or make correct guesses about the countries in question or convergence clubs.” If an incorrect convergence club is examined, the likelihood of detecting convergence is greatly weakened. A separate approach, mostly conduced by economists, divides the sample variance in a given period by the sample mean of this period. Known as a regression approach, it relies on estimating convergence, rather than measuring it. Despite this, Plumper and Schneider (2009) make calls for researching how different approaches could be applied as neither regression nor variance can be completely accurate.

As diffusion research has attracted attention across different social sciences, approaches are highly influenced by a particular scholars disciplinary field. This is in the particular case of the determinants of individual behaviour. Political scientists, on the one hand often emphasise the importance of rationality and self-interest, whereas sociologists may focus efforts on relational ties and organisational structures. Perceptions of where power lies in the diffusion processes may be an issue. For coercion, why some people are recipients whilst others are the ones doing the coercing can be contested as well as the basis of whether a policy implementation was due to a more ‘rational-thinking’ than coercion at all. These assumptions could lead to different findings and difficulty in building diffusion theory. It is also true that most research in the area looks at methods of diffusion having occurred in the past, rather than examining the processes while still in practise. Problems of ‘selective accounts’ of what happened during the policy change process exist and researchers are forced to depend on accounts which are susceptible at under-playing or over-playing the existence of external pressures and influences. In cases of a successful policy implementation, the possibilities of its actual origins may be forgotten and as described by Rose (2005) the policy’s legacy becomes simplified to “the way we do things here”. The issues with research therefore begin with even attempting to uncover the policy transferring process.

Mixed Versions of Rationality

As Bennett and Howlett (1992) note there is little empirical work which “unequivocally demonstrates that X would not have happened had ‘learning’ not taken place.” Despite an application of improved research approaches, it might ultimately be simply “impossible to observe the learning activity in isolation from the change requiring explanation.” It can be argued that many of the work under which diffusion is examined could instead be viewed under a ‘rational decision-making’ framework rather than part of a separate mechanism altogether. Under this perspective, much of the policy learning such as lesson drawing outlined above may simply indicate a rational policy change (James and Lodge, 2003). When Rose (2005) examines such a learning, she acknowledges that it is widespread and increasing. However, it is perhaps difficult to make a distinction or think of “any form of rational policymaking that does not, in some way, involve using knowledge about policies in another time or place to draw positive or negative lessons” (James and Lodge, 2003). A rational preference for also maintaining a status quo in a country, such as Ireland maintaining strict reproductive and bodily autonomy laws for many years despite worldwide norms that differed, could be viewed as involving ‘negative lessons’ about alternatives in other countries. The framework almost associates policymakers with different kinds of rationality towards decision making in different instances. Learning, for example, assumes those in power employ a ‘perfect rationality’ to adjust the goals or techniques of policy considering the consequences of past policy and new information. This is done in order to so better attain the ultimate objects of governance. Emulation, on the other hand, is more difficult to prove as it invokes a lack of or a ‘bounded rationality’ element amongst decisionmakers.

A Political Rather Than Technocratic Process:

The forms of diffusion examined are namely said to spread due to decisionmakers evaluating the policy consequences they observe in other units. Other countries are to be viewed as policy laboratories. However, as noted previously, policy processes are not apolitical and ideas-based and institutional accounts of the actors involved is always worth examining (Bennett, 1991). Within diffusion, there is the need to pay further attention to the problem definition and agenda setting stage of the policy cycle. In the case of social policy within Brazil, ideology and social networks of policymakers were found to be leading reasons on whether policy diffusion through emulation took place (Sugiyama, 2008). Individuals can make decisions about the distribution of resources due to ideology, regardless if it runs counter to their interests. Ideology can “influence political actors in actionable ways, as it filters their information, shapes their worldview, and guides their evaluation of particular policies” (Mullins, 1972). In the case of social programmes in Brazil which would extend services to marginalized and poverty affected groups, the study found that more left learning progressive actors emulated the programmes in their home state. In terms of networks and socialisation, Kaufman (1999) reminds us that human behaviour is embedded in a mix of organisational and informal relationships that can shape our preferences. Looking at policy choice on a national scale, Gilardi (2010) demonstrates that decisionmakers not only learn from policy direct effects, but also from political consequences. This is especially true for electoral consequences and such processes can be tied up with ideological stances. Gilardi and Wasserfallen (2017), for example, imagine that a “conservative governments may strongly rely on evidence from other units suggesting that cuts in welfare spending have no negative electoral consequences and increase economic growth, while ignoring information suggesting the opposite.” There is massive selectivity at play in information processes.

Conclusions:

Proponents of ‘policy diffusion’ argue that it has high salience and as such “when we are analysing policy change, we always need to ask the question: Is policy transfer involved?” (Dolowitz and Marsh, 2000). This paper has classified four forms of diffusion. Non-coercive forms of diffusion occur when a country or region adopts a policy similar to that of somewhere else, without the existence of pressure being exerted to do so. This can take the form of learning, where policymakers look at policies which have been applied to similar problems in different settings and learn from them. Non-coercive transfer can also occur in times of uncertainty when those making decisions are unsure of what to do and what measures to adopt. Coercive forms of diffusion, on the other hand, stem from external pressures put on policymakers to adopt a specific type of policy. This can take the form of a soft pressure competitively as countries compete to maximise resources or it can be due to threats of military intervention, legal threats or financial fines and awards. The conceptual distinctions that diffusion offers supports the claim that it constitutes a class of convergence mechanisms in its own right. The process acknowledges that mechanisms other than harmonisation, imposition or accidental accumulation of reactions to identical problems exist, particularly in the case of competition and emulation.

However, diffusion examined as a simple ‘transfer of policy’ is bound to be misleading, conjuring up an almost straightforward process, when in fact, the reality is very different. I believe that there is little evidence as the literature currently stands that allows us to dissect the aspects of diffusion from the remainder of other potential factors at work. As such, while I do believe the general explanatory potentials of these approaches, if we want to examine policy convergence on a macro-level perhaps a better place to begin is with globalisation, political internationalisation and other modernising forces. Diffusion cannot explain a change of policy at an origin country before the transferring process. During particular examinations of transfer at a smaller scale, I also believe that it is highly important to be aware of the politized nature of convergence. Diffusion recognises this in the form of coercive power, however, other elements such as ideas and structural settings must also be understood. What is ‘chosen’ to be diffused and what is ‘chosen’ not to be is an interesting question to ask. As Bennett has stated, it is “not enough to say that comparable conditions produce comparable problems which produce comparable policies. There are also different political mechanisms, … through which policies might converge” (Bennett, 1991). In most cases mentioned what initially appears to be a wholescale policy learning also involves many other factors. On ‘closer inspection’ a rather loose and not very detailed emulation often takes place with a transfer of attitudes or symbols, rather than complete policies. Policy change must be viewed as a ‘hybrid’ and no single explanatory factor can be used alone to account for such change. Policy diffusion theory, therefore, must only be considered in conjunction with other policy change theory.

Bibliography

Bennett, C. J. (1991). What is policy convergence and what causes it?. British journal of political science, 21(2), 215-233.

Bennett, C., & Howlett, M. (1992). The lessons of learning: Reconciling theories of policy learning and policy change. Policy Sciences, 25, 275-294.

Berry, F.S. and Berry, W.D. (1999) ‘Innovation and diffusion models in policy research’, in P.A. Sabatier (ed.), Theories of the Policy Process, Boulder, CO: Westview Press, pp. 169– 200.

Biesenbender, S. and Tosun, J. (2014). Domestic politics and the diffusion of international policy innovations: How does accommodation happen? Glob. Environ. Change, 29, pp. 424-433.

Braun, D. and Gilardi, F. (2006). Taking ’galton’s problem’ seriously. towards a theory of policy diffusion. Journal of Theoretical Politics, 18(3):298–322.

Browers, M. (2003) ‘The Reconciliation of Political Theory and Comparative Politics’, in J.S. Holmes (ed.), New Approaches to Comparative Politics: Insights from Political Theory (Oxford: Lexington Books): 7–22.

Darby, J. and Mac Ginty, R. (2001). Guns and Government: The Management of the Northern Ireland Peace Process, Springer.

DiMaggio, P.J. and Powell, W.W. (1983) ‘The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields’, American Sociological Review, 48(2): 147–60.

Dobbin, F., Simmons, B., & Garrett, G. (2007). The global diffusion of public policies: Social construction, coercion, competition, or learning? Annual Review Sociol., 33, 449-472.

Dolowitz, D. P. (1997). British employment policy in the 1980s: learning from the American experience. Governance, 10(1), 23-42.

Dolowitz, D. P., & Marsh, D. (2000). Learning from abroad: The role of policy transfer in contemporary policy‐making. Governance, 13(1), 5-23.

Drezner, D. W. (2001) ‘Globalization and policy convergence’, International Studies Review 3: 53– 78.

Eyestone, R. (1977). Confusion, diffusion, and innovation. American Political Science Review, 71(2), 441-447.

Finnemore, M. (1996) ‘Norms, culture, and world politics: insights from sociology’s institutionalism’, International Organization 50(2): 325 –48.

Finnemore, M. and Sikkink, K. (1998) ‘International norm dynamics and political change’, International Organization 52(4): 887 – 917.

Finnemore, M. and Sikkink, K. (2001). “Taking Stock: The Constructivist Research Program in International Relations and Comparative Politics.” Annual Review of Political Science 4:391–416.

Genschel, P. (2002). Globalization, tax competition, and the welfare state. Politics & Society, 30(2), 245-275.

Gilardi, F. (2012). Transnational diffusion: Norms, ideas, and policies. In Carlsnaes, W., Risse, T., and Simmons, B. A., editors, Handbook of International Relations, chapter 18, pages 453–502. SAGE Publications, London.

Gray, V. (1973) ‘Innovations in the state: a diffusion study’, American Political Science Review 67(4): 1174 –85.

Hall, P. A. (1993). Policy paradigms, social learning, and the state: the case of economic policymaking in Britain. Comparative politics, 275-296.

James, O. and Lodge, M. (2003) ‘The Limitations of “Policy Transfer” and “Lesson Drawing” for Public Policy Research’, Political Studies Review, 1: 179–93.

Jordan, A., Wurzel, R. Zito, A. and Brückner, L. (2003) ‘European Governance and the Transfer of “New” Environmental Policy Instruments (NEPIs) in the European Union’, Public Administration, 81(3): 555–74.

Lazer, D. (2005) ‘Regulatory capitalism as a networked order: the international system as an informational network’, The Annals of the American Academy of Political and Social Science 598 (March 2005): 52 – 66.

Levi-Faur, D. (2005) ‘The global diffusion of regulatory capitalism’, The Annals of the American Academy of Political and Social Science 598 (March 2005): 12 – 34.

Levi-Faur, D. and Jordana, J. (2005) ‘Preface: The making of a new regulatory order’, The Annals of the American Academy of Political and Social Science 598 (March 2005): 6 –9.

Mullins, W. (1972). On the concept of ideology in political science. American Political Science Review, 66(2), 498-510.

Page, E.C. and Mark-Lawson, J. (2007) ‘Outward-Looking Policy Making’, in H. Bochel and S. Duncan (eds), Making Policy in Theory and Practice (Bristol: Policy Press): 47–64.

Plümper, T., & Schneider, C. J. (2009). The analysis of policy convergence, or: how to chase a black cat in a dark room. Journal of European Public Policy, 16(7), 990-1011.

Porter, G. (1999). Trade competition and pollution standards:“race to the bottom” or “stuck at the bottom”. The Journal of Environment & Development, 8(2), 133-151.

Rose, R. (2005). Learning from Comparative Public Policy. A Practical Guide, London: Routledge.

Shipan, C. R. and Volden, C. (2008). The mechanisms of policy diffusion. American Journal of Political Science, 52(4):840–857.

Simmons, B. A., Dobbin, F., and Garrett, G. (2006). Introduction: The international diffusion of liberalism. International Organization, 60:781–810.

Simon, H. A. (1991). Bounded rationality and organizational learning. Organization science, 2(1), 125-134.

Sugiyama, N. B. (2008). Theories of policy diffusion: social sector reform in Brazil. Comparative Political Studies, 41(2), 193-216.

Swank, D. (2006). Tax policy in an era of internationalization: Explaining the spread of neoliberalism. International Organization, 60(4), 847-882.

Tanguay, G. A. (2001). Strategic environmental policies under international duopolistic competition. International Tax and Public Finance, 8(5-6), 793-811.

Tran, M. (2003). France and Germany evade deficit fines, The Guardian, 25 Nov 2003, online Available at: https://www.theguardian.com/business/2003/nov/25/theeuro.politics Accessed 11/05/19.

Walker, J.L. (1969) ‘The diffusion of innovations among American states’, American Political Science Review 63(3): 880– 99.

Wasserfallen, F. (2017). Policy Diffusion and European Public Policy Research. In The Palgrave Handbook of Public Administration and Management in Europe. Basingstoke: Palgrave Macmillan.

Weyland, K. (ed.) (2000) Learning from Foreign Models in Latin America Policy Reform (Washington, DC: Woodrow Wilson Center).

Wheeler, D. (2001). Racing to the bottom? Foreign investment and air pollution in developing countries. The Journal of Environment & Development, 10(3), 225-245.

Zhou, S., Matisoff, D., Kingsley, G. and Brown, M. (2019). Understanding renewable energy policy adoption and evolution in Europe: The impact of coercion, normative emulation, competition, and learning, Energy Research and Social Science, 51: 1-11