Theory of Policy Learning, China
Policy learning is a common learning process whereby knowledge, policies, or administrative arrangements shift from one nation or policy domain to another.
The earliest formal identification by the Western world of policy learning is that of Aristotle in his Nicomachean Ethics in which he recommended that the Polis should engage in drawing positive and negative lessons from the development of great city states (see Evans and Marsh 2012). By the end of the twentieth century, scholars such as Common (2004), Carroll and Common (2013), Rose (2005), Evans (2012), and Dolowitz and Marsh (1996) have used the concept of policy transfer as an umbrella term for defining a range of processes of learning which integrates different policy elements from different types of organization. Indeed, policy learning theory has increasingly been used as a practical tool for affecting “quick fix” and “short cut” solutions to policy problems.
Debating Policy Transfer
There is nothing new about the concept and application of policy-oriented learning, policy transfer, and lesson-drawing. By the end of the twentieth century, scholars such as Rose (2005), Dolowitz and Marsh (1996), Evans (2012), and Carroll and Common (2013) have used the concept of policy learning (or derivatives such as policy transfer and lesson-drawing) as umbrella terms for understanding a process in which knowledge about institutions, policies, or delivery systems at one sector or level of governance is used in the development of institutions, policies, or delivery systems at another sector or level of governance (Evans and Marsh 2012). Different forms of policy transfer such as bandwagoning (Evans and Marsh 2012) have been identified in a wide ranging and evolving literature which has attracted significant academic attention from domestic (Dolowitz and Marsh 1996), comparative (Wolman 1992), and international political scientists (Stone 2000), as well as public management experts (Common 2004; Carroll and Common 2013).
Policy learning and its derivatives are used as a container concept for quite different claims about why public organizations engage in this form of policy development. Why has there been an upsurge of interest in policy learning? There are at least six main reasons for this. First, processes of globalization both external to the nation state (e.g., through changes in the nature of geopolitics, political integration, the internationalization of financial markets and global communications) and the “hollowing-out” of the nation state itself have created new opportunity structures for policy transfer. There are four interrelated trends which illustrate the reach of this process: privatization and limiting the scope and forms of public intervention; the loss of functions by central government departments to alternative service delivery systems (such as special agencies) and through market testing; the loss of functions from the British government to European Union (EU) institutions; and the emergence of limits to the discretion of public servants through the new public management, with its emphasis on managerial accountability, and clearer political control created by a sharp distinction between politics and administration. One further dimension can be added to these four – the global trend towards regionalization and devolution.
Second, policy learning is more likely to occur in an era of governance with the increasing pervasiveness of “self-organizing, inter-organizational networks,” which compliment markets and hierarchies as governing structures for authoritatively allocating resources and exercising control and coordination. In times of uncertainty, policy-makers at the heart of networks will look to the “quick fix” solution to public policy problems that policy transfer can provide either through other governments or from other sectors (e.g., the Community Sector and social inclusion policy). In a sense, we can understand policy learning as evidence based policy-making on the hoof.
Third, ideological symmetry between governments provides a strong foundation for policy-oriented learning. For example, processes of learning across time and space characterized the flow of ideas between the Clinton, Blair, Rudd, and Gillard governments in the UK, the USA, and Australia particularly in the area of social policy reform. Prominent examples include: education (reduction of class sizes), crime (zero-tolerance, antitruancy drives), and welfare reform (welfare to work and creation of work incentives).
Fourth, for policy scientists it provides a rich source for investigating structures and processes of collaborative governance as policy transfer is often the outcome of collaborative activities between governmental and nongovernmental actors which have become increasingly salient in an era of governance.
Fifth, it is evident that structures of collaborative governance such as those that emerge to steer processes of policy learning can both extend and limit participation; empower and disempower affected societal groups. We shall see a little later that policy learning tends to be reserved for elite groups and actors. Only the select few can join in!
The sixth main reason why there has been an upsurge of interest in policy learning is that the securing of loans by developing countries from the International Monetary Fund and elsewhere has become conditional on the introduction of adjustment programs including initiatives such as privatization. Such programs are predicated on a western interpretation of “Good Governance” and give rise to negotiated processes of policy transfer.
The modern study of policy learning originates from a subset of the comparative politics literature; policy diffusion studies. Research in this area focused on identifying trends in timing, geography, and resource similarities in the diffusion of innovations between countries (and, in the USA, between states). However, the research revealed next to nothing about the content of new policies. In contrast, the contemporary literature on policy learning primarily focuses on voluntary learning between jurisdictions, identifying a process in which policies implemented elsewhere are examined by rational political actors for their potential utilization within another political system. More recently the literature has begun to address questions concerning coercive learning. As Rose (2005) puts it, “Every country has problems, and each thinks that its problems are unique…However, problems that are unique to one country…are abnormal…confronted with a common problem, policy makers in cities, regional governments and nations can learn from their counterparts elsewhere responded.”
In sum then, the literature on policy learning analysis may be organized into two discernible schools: one which uses the label “policy transfer” but deals with different aspects of a process of policy-oriented learning or lesson-drawing using different nomenclature; and one which uses the concept directly. The former includes the literatures on: bandwagoning (Ikenberry 1990); convergence (Bennett 1991); diffusion (Majone 1991); evidence-based practice; learning (Greener 2001); and lesson-drawing (Rose 2005). The latter includes the work of domestic policy scientists (Dolowitz et al. 2000; Evans and Marsh 2012; Page 2000), comparativists’ (Common 2004; Wolman 1992); and international policy scientists (Carroll 2013; Haas 1992; Stone 2000).
There are four main approaches utilized in policy learning analysis: policy transfer analysis, personal interaction approaches, policy diffusion studies, and inclusive approaches.
Policy Transfer Analysis
Policy transfer analysis is used by policy scientists who focus on the process of policy transfer directly in order to explain the voluntary, or negotiated importation of ideas, policies or institutions. It has tended to be a predominantly interstate approach which emphasizes the role of state actors as active agents seeking solutions to policy problems rather than the passive agents depicted in pluralist or corporatist decision literatures. The role of the agent of transfer (e.g., policy entrepreneurs such as think tanks, knowledge institutions, or pressure groups) is also considered to be critical within this approach.
Two key applications of the concept can be identified within this approach. Bennett (1991) utilizes policy transfer as his key independent variable. In this sense, the process of transfer is used to explain why a particular policy was adopted. For example, Bennett’s work on policy convergence focuses on the transfer of policy goals, content, and instruments (e.g., privatization programs). He has conducted empirical research on, among other areas, how and why evidence about the United States’ Freedom of Information Act (FOIA) was utilized in Canada and Britain. He reveals that FOIA was used as an exemplar in Canada and the reverse in Britain.
Policy convergence reveals a process in which international solutions are implemented to solve domestic state problems. Bennett (1991) claims that the word “convergence” refers to the phenomenon of Policy A existing in many countries. He identifies three main phases of policy transfer: (1) the awareness of the program, (2) the utilization of knowledge, and (3) the adoption of the same program (Bennett 1991). In short, Bennett’s policy convergence (1991) approach suggests that evidence of success in policy, administrative arrangements, and institutions in Country A leads to the adoption of policy, administrative arrangement, and institutions in Country B. By implication studies of policy convergence focus on structural factors such as policy outcomes as the products of institutions.
A second usage of the term can be found in the work of Richard Rose (2005). Rose uses policy transfer as a dependent variable. He attempts to explain why a policy transfer occurs and how lessons are incorporated into a political system through various processes of copying, emulation, hybridization, or inspiration. Rose (2005) emphasizes the role of the bureaucrat and the program itself.
If formal approaches focus on the structural aspects of policy learning rooted in institutional analysis, ideational approaches focus on agency. This approach argues that policy learning is almost solely based on the personal interaction between agents of transfer, bureaucrats and politicians within interorganizational settings. As Evans (2012) notes, “the study of policy transfer analysis should be restricted to action-oriented intentional learning: that which takes place consciously and results in policy action.”
In these settings, there exists a pattern of common kinship expressed through culture, rules, and values. Hence, an emphasis is placed on analyzing the structure of decision through which policy transfer takes place and relationships between agents of transfer and their dependencies. Here ideas matter. For example, Carroll and Common’s (2013, p. 4) impressive Policy Transfer and Learning in Public Policy and Management focuses on “the need to consider the processes within which policy ideas, techniques and practice are exchanged.” Stone’s work (2000), identifies think tanks as key agents of policy transfer within what are termed “epistemic communities.” Epistemic communities are comprised of natural and social scientists or individuals from any discipline or profession with authoritative claims to policy relevant knowledge that reside in both national, transnational and international organizations. The function of these communities is to facilitate the emergence of policy learning that may lead to policy convergence. The epistemic communities’ literature has been used to explain how international policy has converged in areas such as GATT, food aid, financial regulation, and environmental issues through regime politics.
The social learning approach is also associated with the personal interaction approach. It focuses on the factors which influence how politicians and policy-makers learn how to learn and addresses problems such as when and how politicians, policy-makers, and societies learn how to learn. An emphasis is placed on seeking to understand the psychology of individual and group behavior as a trade-off between organizational and societal influences.
Policy Diffusion Studies
A distinct categorization for policy diffusion studies may be identified because such studies, while not using the concept of policy transfer directly, seek to identify and quantify patterns of diffusion that are germane to the process of policy transfer itself. Such studies are common in the USA in relation to the diffusion of innovations across state jurisdictions (Weyland 2007) and local governments (e.g., Shipan and Volden 2008) in European Union studies (see Füglister 2012) and in international studies.
Peters (1997), for example, examines the diffusion of administrative reform policy transfers through the member countries of the OECD. He argues that policy transfer, or policy learning, is a common activity in governments around the world, but that there are differences in the rates at which countries are able to learn and adapt. These differences are explained by structural factors such as economic, ideological, cultural, and institutional commonalties. Those countries similar in these ways are more likely to transfer policy. He concludes that cultural variables play an extremely important role in the transfer of policy innovations among countries particularly in relation to geographical proximity and political similarity. However, another set of policy ideas, those associated with political parties and ideologies, appear to have much less of a relationship with the spread of managerialist reforms.
Dolowitz and Marsh (1996, 2000) lead efforts within British political science to develop an inclusive model of policy transfer which draws on the work of Rose (2005), Bennett (1991), and Wolman (1992). In essence, they draw together a general framework of heterogeneous concepts including policy diffusion, policy convergence, policy learning, and lesson-drawing under the umbrella heading of policy transfer. Hence, policy transfer is used as a generic concept which encompasses quite different claims about the nature of policy development. Dolowitz and Marsh (1996, p. 357) provide an extremely useful framework which invites others to criticize and develop it, “[w]e have suggested a series of questions which can be used both to organise our current knowledge of the process and to guide future work.” The framework developed by Dolowitz and Marsh is clearly designed to incorporate a vast domain of policy-making activity by classifying all possible occurrences of transfer, voluntary and coercive, temporal and spatial. Policy transfer is common in this scheme and processes such as the “rapid growth in communications of all types since the Second World War” have accelerated the process (1996, p. 343). This is not, however, a claim that policy transfer is all-pervasive and Dolowitz and Marsh make no comment on the difference between policy transfer and normal forms of policy innovation, nor whether policy transfer falls entirely within the parameters of policy succession.
Understanding the Process of Learning and Globalization
Firstly, in all of the cases processes of globalization (e.g., WTO membership, global communications, adoption of international financial, property, strategic, or public management system) provided opportunity structures for learning to take place. There is no doubt that international organizations have had a growing influence in shaping domestic reform agendas. It is therefore no coincidence that the same discourses in terms of social, economic, and administrative reform can be found in completely dissimilar societies. This is due to a large extent to what Evans (2012) identifies as “… a political process of global institution building…” that underpins the economic dynamic of globalization. But not only financial international organizations have a say in the definition of the reform agenda; international knowledge organizations such as the WTO also have a determinant role in the agenda setting process. However, it would be wrong to conclude that the increased internationalization of reform practices has limited the development of indigenous content in reform policies. As we have seen the reform processes in this study have not slavishly followed the international benchmark. In most of the cases, initially copying took place but then in the implementation process the objects of transfer were combined with indigenous lessons.
Secondly, it is equally clear that the political orientation of a country does not determine the adoption of certain reform ideas. Governments emanating from either left or right of the political spectrum are equally likely to be involved in introducing neoliberal market reforms.
Thirdly, in all of the cases, processes of social (demographic and cultural) change arising from economic reform (marketization – financial services, securities, real estate, social security) also provided opportunity structures for learning to take place. However, they also exposed fundamental sources of social antagonism between Party elites seeking to protect their privileges and reforming liberal elites striving for better policy. This observation was particular acute in the areas of banking, property management, and growth enterprises reform.
Fourthly, in all of the cases modernization processes also occurred due to citizen dissatisfaction with the existing quality of provision and increased expectation for better services. However, in all sectors of reform the Chinese government lacked expertise to embed new working norms and values and were therefore reliant on technical support from international organizations or overseas governments.
Fifthly, all of the cases represented a significant ideological challenge to the status quo and by implication provide a site of struggle between different conceptions of the public good – party accountability versus public accountability, communal property versus private property, state responsibility versus individual responsibility, and state monopoly capitalism versus neoliberalism. Hence, in most cases we see the reemergence of old elites in new class positions – former party bosses become property magnates.
Agents of Policy Transfer
There are five observations about the work of these agents of learning that need to be highlighted here. Firstly, the utilization of the expertise of leading international knowledge and financial organizations has helped to legitimate the reform processes themselves. Policy-makers do not only turn to advanced countries experiences but to the guidelines and recommendations formulated by international bodies such as the WTO, IMF, the WB, OECD, and the UN among others (see also Ladi 2004 and Stone 2000). These organizations are increasingly important in shaping the character of transition states as they establish what is considered to be “best practice” in reforming the economy and the public sector. Certain international organizations have the necessary coercive means to force developing countries to adopt certain policies (e.g., the IMF and the WB) and others use moral coercion to pressurize countries to embrace certain reforms (e.g., OECD). In the case of China, the government clearly has greater autonomy in determining its own path to development although it needs to make certain concessions to international benchmarks along the way. Moreover, as international organizations are considered to be authorities in their particular area of expertise, policies, and programs based on their guidelines and principles have implicit international recognition of being in line with best practice.
Secondly, the main internal agents of learning are members of the Chinese bureaucratic elite who share a commitment towards the same policy objectives and have common norms and values. It is evident that this elite is committed to modernization and is relatively cohesive.
Thirdly, it is notable that in all our cases formal and informal institutional structures for policy learning including internal and external agents of transfer were created to facilitate policy change whether it be a sophisticated policy hub.
Fourthly, the policy learning is characterized by the development of long-term relationships with external agents of transfer.
Fifthly, in all of our cases cohesive policy networks were created leading to fast track processes of learning. However, social conflict emerged in the implementation processes exposing a war of ideas between conservative party elites and neoliberal reformers.
Barriers to Transfer
There are three broad sets of variables can be identified in the case study literature on policy transfer: “cognitive” barriers, “institutional” barriers, and “environmental” barriers (Evans 2012). These variables interact in complex and often unexpected ways and inform the process of policy transfer.
“Cognitive” barriers refer to the process by which public policy problems are recognized and defined in the predecision phase, the rationality of the search conducted for ideas, the receptivity of existing policy actors and systems to policy alternatives, and the complexity of choosing an alternative. The most significant cognitive barriers for agents of policy change to overcome at this stage of policy development are the prevailing organizational culture and the need for effective cultural assimilation. In my understanding, cognitive obstacles in the policy formulation phase of the process of policy transfer constrained policy development due to the absence of a genuinely evidence-based approach to evaluating the compatibility of the objects of transfer. Moreover, cognitive obstacles in the implementation phase of the process of policy transfer led to resistance from party elites until a strategy of elite renewal was crafted. In terms of institutional barriers, the absence of technical capacity to deliver the programs meant that an incremental approach to implementation took place underpinned by capability development programs. Moreover, certain reforms were characterized by organizational resistance to change at the middle management level due to the need for new skills, norms, and values to affect organizational change.
“Institutional” barriers refer to the normal technical implementation constraints that constrain or facilitate the process of lesson-drawing. This includes coherent and consistent objectives; the incorporation of an adequate causal theory of policy development; the sensible allocation of financial resources; hierarchical integration within and among implementing organizations; clear decision rules underpinning the operation of implementing agencies; the recruitment of program officers with adequate skills/training; sufficient technical support; and the use of effective monitoring and evaluation systems including formal access by outsiders.
“Environmental” obstacles refer to the absence of effective cognitive and elite mobilization strategies deployed by agents of policy change, the need for the development of cohesive policy transfer networks to ensure successful policy-oriented learning, the broader structural constraints (institutional, political, economic, and social) that impact and shape the process of lesson-drawing.
Environmental obstacles during the process of policy transfer constrained policy development particularly resistance from the Party until a strategy of elite renewal was crafted which led to the use of the nomenklatura system to affect control over the process of policy development. This suggests that it is crucial to build a societal consensus across elites on program imperatives in order to ensure the effective implementation of new programs.
How can we judge the success of these interventions? David Marsh and Allan McConnell (2010) provide us with the basis of a framework for evaluation. They develop a heuristic device for assessing success, focusing on three dimensions: process success; programmatic success; and political success. Process success is defined as “the stages of policy-making in which issues emerge and are framed, options are explored, interests are consulted and decisions made” (Marsh and McConnell 2010, p. 572). Central to Marsh and McConnell’s conception of process success is the notion of legitimacy; understood here in terms of due process and democratic values. Innovation and influence can also be measures of process success, irrespective of the particular policy outputs and outcomes. Innovation can also be based on policy transfer from another political jurisdiction (see: Dolowitz and Marsh 2000).
Programmatic success refers to the assumption that “programmatic” success is more likely if the policy process involves, and reflects the interests of, a sufficiently powerful coalition of interests (Marsh and McConnell 2010, p. 573). “Programmatic” success is often seen as synonymous with policy success and the contemporary focus among most Western democracies on evidence-based policy-making. And, finally Marsh and McConnell understand political success from two perspectives – from the perspective of government and civil society. For government, a policy may be successful if it assists their electoral prospects, reputation, or overall governance project. However, perceptions of the success of a policy will differ depending on your social location. This requires some reflection on the question: success for whom? As Bovens et al. (2006, p. 322) argue: uncontroversial instances, policy evaluations are entwined with processes of accountability and lesson-drawing that may have winners and losers. However, technocratic and seemingly innocuous, every policy program has multiple stakeholders who have an interest in the outcome of an evaluation: decision makers, executive agencies, clients, and pressure groups.
Policy Learning in China
Theoretically, the previous analysis of policy learning termed as policy transfer and lesson-drawing, such as Rose (2005), Bennett (1991), Dolowitz and Marsh (1996) and others, has tended to concentrate on the approach of transferring policy, the method of the transfer, and the lessons derived from it (see: Evans and Marsh 2012). However, there is also an increasing interest in exploring what elements influence the success of policy learning (see Marsh and Sharman). This puzzle has become particularly salient with regard to the role of the Chinese state in steering a highly effective economic reform project in which policy learning appears to have played a fundamental role promoting the question – is China particularly adept at policy transfer and policy-oriented learning and if so why?
Chinese policy-makers are keen to learn about what works from overseas countries that they admire. It is noteworthy that Chinese interest in policy transfer is said to have predated the West. As early as 422 BC, the first historian Qiu Zuo, conjured with the idea of lesson-drawing as an idiom as “yin yiweijian” in his book named Guo Yu. The idiom of “Yin Yi Wei Jian” translated as “The overturning of the cart in front is a warning to the carts behind,” which means a person or organization should be rational enough to learn positive and negative lessons from others. Fifty years later in 372 BC, the Chinese sage Mencius advocated “yīnjiànbùyuǎn” (Hulsewé 1979), which means you can learn lessons regardless of the space and time period. And the most famous Chinese philosopher Confucius also said: “line three, there must be someone better than me, learn the good from the one to replace my weakness, and avoid the negative lessons.” These extracts from Chinese literature demonstrate a long history of lesson-drawing as a theory of praxis in Chinese policy-making.
In the course of the last three decades, processes of globalization (economic integration, geopolitics, communications, political integration) have created a great trading-pit in policy ideas with “ready-made”, “off the shelf” solutions to policy problems. This supermarket approach to policy learning has inhibited rational policy learning in which policy-makers use credible instruments of policy analysis such as evidence-based evaluation methods to assess the costs and benefits of pursuing a policy drawn from a nonindigenous setting. The Chinese government has been particularly vulnerable to this approach as the “Reform and Open Door Policy” has provided a perfect environment for the application of “foreign” models to ease transition problems. Methods of adaptation, emulation, hybridization, synthesis and inspiration have played an integral role in economic and social reform. Although the results have often been impressive, there has been a rising concern among academics and practitioners about why China has been so quick to learn lessons from Western exemplars rather than steer its own course or learn from states at a similar state of development.
Policy Learning and the Rise of the New China
Since the implementation of “Reform and Open Door Policy” at the early 1980s, China has experienced a great and historical social change due to a progressive incremental reform that has delivered a result of shifting the country from a command to a market socialist economy and latterly to what may be termed a new authoritarian development state. Indeed, in order to meet the perceived imperatives of state transformation, Chinese government has launched a series of profound reform in its government sector and public institution, programmed to promote economic and social development, this has been termed the “reform and Open Door” policy has largely been facilitated through processes of policy learning.
The growth of policy learning has become an inevitable trend in Chinese administrative and policy development due to the increase of global economic pressures, the rapid internal growth communications technology and the increasing influence of international organizations in setting international policy agendas (Larmer 2005). Processes of economic globalization have galvanized the Chinese government into becoming more and more active in incorporating the policy experiences of developed countries into its domestic reform program. This is in keeping with two overlapping claims about the centrality of institution-building in development which have gained apostles in Chinese government in the past decade and which have had an important influence on institutional reform in particular. The first claim is that institutions matter in development because they provide stable governing parameters (Leftwich 2006); the second claim is that the achievement of good governance in “Chinese-style” public administration ensures that institutional rules remain both stable and effective. Hence, the deepening and consolidation of the reform of the administrative system is perceived by party elites to be of vital importance to ensure the continuation of the success of the economic reform project in China (Evans and Marsh 2012). Furthermore, there has been a sharp focus on New Public Management (NPM) type reforms, that is, the application of private sector methods and public choice precepts in the management and delivery of public goods. It is also observed that international cooperative programs have become the key instrument of policy learning in this area.
Processes of Policy Learning in China
The cumulative effect of the interaction of pressures and ideas emanating from dynamic international and national structures and their interpretation by different agents create the necessary conditions for policy transfer to take place. In China, processes of opening-up, elite renewal and international pressures to conform are some of the myriad explanatory variables that prompted policy-makers to search overseas for solutions to the problems they are facing. However, it would be wrong to conclude that the increased internationalization of reform practices has limited the development of indigenous content in reform policies. As China have seen the reform processes in this study have not slavishly followed the international benchmark. In most of the cases, initially copying took place but then in the implementation process the objects of transfer were combined with indigenous lessons. In short, learning through doing occurred in the process of implementation. This demonstrates the pragmatic nature of Chinese policy learning in keeping with the Chinese political tradition. As Sitwell’s observation so acutely observes, the genius of the Chinese national culture lies in its ability to assimilate overseas ideas and adapt them to their own needs and desires. It also highlights the interactive nature of structure and agency. Here the essential components of the national culture are preserved within the policy learning design and inappropriate foreign content is filtered out.
Besides, promoting an “Economy, Efficiency, and Effectiveness” administrative system must be prime objective in the economic transition. As Turner and Hulme (1997, p. 1) observe, “It seems that all are agreed on the proposition that the nature and performance of public sector organizations are critical elements in determining developmental success.” An successful administrative reform requires to build a strong, efficient and effective central administrative system which aims at improving efficiency, effectiveness and responsiveness. Indeed, countries such as South Korea, Singapore, Thailand, and Malaysia all have been witnessed by the most rapid and sustained development regarding to the highly effective public sector organizations, strong centers, entrepreneurial elites, relatively autonomous states, effective economic planning ministries, certain aspects of good governance, and have engaged in civil service reform (see Leftwich 2006).
Furthermore, the successful economic transition and administrative reform require engaging in progressive lesson-drawing and policy-oriented learning from international experience but that ultimately indigenous solutions must be found to indigenous problems. Public administrators must engage in international comparison and rational policy learning and draw on the most proper overseas expertise that fit their own circumstances.
In all of our cases long implementation processes were necessary to achieve effective outcomes. It is also noteworthy that public organizations are normally unlikely to engage in policy transfer if the policy is too complex and difficult to implement (Evans and Marsh 2012). However, in these cases, which all involved complex transfers, the Chinese government moved quickly to copy from overseas exemplars in the policy formulation process and then to hybridization in the implementation process. This is wholly in keeping with Chinese learning traditions and its emphasis on pragmatism and was also reflected in the dominant view that “state of development” issues were not central to evidence-based thinking in policy transfer. Selecting the most appropriate exemplar at a similar state of development is considered fatalism. Hence “to be the best you have to learn from the best.”
Measuring Success in China
The heuristic requires slight modification to each of the three dimensions to operationalize it effectively in the Chinese context. Firstly, in terms of process success, the concept of legitimacy as applied to liberal democracies is underpinned with reference to due processes of constitutional and quasi-constitutional procedures and values of democracy, deliberation, and accountability. This would not be the case in China where the notion of legitimacy in the formation of choices would be reproduced through due processes or procedures underpinned by Party norms and values. Moreover, reference to the passage of law rather than legislation would be more culturally apposite. In addition, given that the heuristic is being applied to a process of cross cultural learning; it makes sense to evaluate success in terms of how cohesive the policy transfer network was in facilitating the learning process (see: Marsh and Evans 2012). Moreover, it also makes sense to evaluate the effectiveness of the formal and/or informal institutions which were created to facilitate change.
Secondly, with reference to programmatic success, the importance of issues of rationality and policy compatibility suggest the need to also evaluate programmatic success in terms of whether the policy proposal was informed by the best evidence, whether it was relevant to China’s state of development and, for whom the policy was successful or not.
Finally, in terms of policy success, the issue of government popularity remains an important proxy measure of success in terms of solidifying support for the “One China project” and in maintaining support for dominant party factions rather than in formal liberal democratic elections. In addition, the issue of international opinion is also important. If the policy is aligned with international benchmarks of best practice, it is more likely to be perceived a success regardless of its social impact.
Avoiding the “Learning Paradox”
The empirical findings identified above provide practitioners with the constituent elements of a better practice framework for guiding processes of policy learning and by implication helping them to avoid what Mark Evans (2012) has termed “the learning paradox.” In other words, lesson-drawing can be a rational and progressive learning activity but only if the program that is transferred is compatible with the value system of the recipient organization, based on the best possible evidence, culturally assimilated through comprehensive evaluation, and, builds on existing organizational strengths. It suggests, following Marsh and McConnell (2010), three dimensions of success – process, programmatic, and political – but with additional indicators that provide a more comprehensive understanding of the learning process. In particular, our understanding of process success is enriched through recognition of the importance of agents (policy transfer networks) in creating ideational structures and diffusing innovative practices and that issues of institutional design are crucial to the generation of successful policy outcomes. Moreover, our understanding of programmatic success is advanced through developing evaluative tools for understanding what can and cannot be transferred (policy compatibility), what the key obstacles to effective policy transfer may be to ensure implementation and in whose interest the program has been developed. And, finally our understanding of political success is progressed through recognition of the importance of elite and cognitive mobilization strategies in winning the war of ideas. In short, evidence is not a sufficient criterion for affecting policy change; politics matters and communicative strategies are central to the achievement of policy success.
Parting Shot of Policy Learning
Each of these applications has its merits. They are sound in their descriptive understanding of policy development and can be used to explain certain aspects of the process of learning. The existing research tells us who has relationships with whom and it can describe how these relations impinge on the making of policy (e.g., why some actors are central and others are not). However, most accounts lack the methodological tools necessary to provide a rigorous understanding of the movement of ideas from one setting to another. Moreover, their understanding of the process of policy learning are at best patchy and are vulnerable to the criticism that it is questionable that transfer has actually occurred at all. As Evans (2012) notes “it may be argued that policy transfer and by implication lesson-drawing does not advance an explanatory theory of policy development,” or, as Ed Page (2000, p. 12) puts it, the researcher should not expect from the literature “firm guidance about how to frame the research questions or how to pursue them empirically” (see also: Dolowitz and Marsh 2000; Evans and Davies 1999). For example, the lesson-drawing and ideational approaches cannot provide a general theorization of policy change that accounts for all processes of lesson-drawing. It can only provide a partial account of policy change emanating from the relationship between the structure of the decision network and the agents operating within them and the network and the policy outcome. If applied in isolation from macro-analysis, this overlooks the potential for the existence of a causal relationship between the network and the environment in which it operates. Moreover, at the same time, the network is interpreted, reinterpreted, and constrained by participating actors. Hence, the complex, interactive relationship between network (structure) and agency needs to be analyzed in any study of policy learning. Moreover, finding the evidence that a lesson has been drawn demands excellent access to key informants in informal decision-making processes. Such access is not often possible either in developed or developing countries.
Two further shortcomings are evident. The first is that the study of policy learning would benefit from a reflection of how traditional organizations can become learning organizations. Rose (2005, pp. 104–5) himself argues that the strategic direction of public organizations are path dependent and characterized by “inheritance rather than choice” in the sense that “past commitments limit current choices.” Hence a set of recommendations on how to break from the “wicked context” problem would be extremely useful (see: Common 2004). Secondly, a more detailed identification of potential obstacles to successful policy learning would provide important insights for practitioners into how to develop: (a) the type of learning organization conducive to the facilitation of successful lesson-drawing and (b) how to develop a model of prospective evaluation to guide effective lesson-drawing.
The Way Forward
In summary then, what would be the properties of a rigorous approach to the study of policy learning? It is proposed here that an approach which: (1) is sensitized to the structure and agency problem; (2) recognizes that issues of institutional design (i.e., how policy networks are designed) are crucial to the achievement of effective policy outcomes; and (3) develops evaluative tools for understanding: (a) what can and cannot be learnt (policy compatibility) and (b) the key obstacles to effective policy learning as a method of prospective policy evaluation provides the way forward.
The Structure/Agency Problem
Simply put, formal transfer analysis and the ideational approach tend to privilege either structure or agency factors in processes of policy transfer. Following Marsh and Smith’s (2000) critique of policy network analysis, I seek to confront this problem by combining the strengths of formal transfer analysis and the ideational approach and integrating them into a coherent analytical approach. The new element which we can take from this approach is that policy transfer processes are driven by policy transfer networks that generate policy outcomes emerging from three interactive or dialectical relationships between: the structure of the network and the agents operating within them; the network and the context within which it operates; and the network and the policy outcome. It is also argued that macrolevel variables are interpreted by both actors and through network relationships and should not be seen as distinct from the object of enquiry – the process of policy learning.
As noted by Evans (2012) policy transfer should be understood as action oriented intentional learning – that which takes place consciously and results in policy action; hence the focus should be on the study of agency. This definition locates policy transfer as a potential causal phenomenon – a factor leading to policy convergence. Nonetheless, policy transfer should be distinguished from policy convergence in that the latter may occur unintentionally – for example, due to harmonizing macroeconomic forces or common processes. The element of intentionality in this definition of policy transfer makes both an agent and ideas essential to both voluntary and coercive processes. Intentionality may be ascribed to the originating state/institution/actor, to the transferee state/institution/actor, to both, or to a third party state/institution/actor.
Designing Instruments of Policy-Oriented Learning
Recent studies of processes of policy learning have emphasized the role of policy networks as key instruments of policy-oriented learning (Evans 2012; Ladi 2002). These are collaborative decision structures comprised of state and nonstate actors that are set up with the deliberate intention of engineering policy change. It is argued that policy networks matter because they shape the nature of policy outcomes emerging from the process of learning. Moreover, the creation of a policy network provides an opportunity structure for the creation of further policy networks.
Evaluative Methods: Determining the Success or Otherwise of Processes of Policy Learning
International organizations, knowledge institutions, and think tanks play a key role in facilitating policy-oriented learning through imparting technical advice, for the content of policy transfers normally reflect areas where indigenous state actors lack expertise. This observation has particular resonance in transition countries such as China as the content of policy transfers often takes the form of conditions imposed by, for example, the WTO in return for membership. Hence, transition countries can become dependent on the technical expertise and other resources of nonstate actors. Agents of policy transfer that can bridge the indigenous knowledge gap can become important players in policy transfer networks. However, the degree of autonomy that state actors can demonstrate from the expertise of nonstate actors tends to be broadly representative of their state of development. For example, in most instances, voluntary forms of policy transfer demonstrate the relative autonomy of state actors in decision processes, while coercive forms of policy transfer demonstrate the incapacity of a state to maintain its national sovereignty over decision-making. Although in the Chinese context, the government has made concerted efforts to guard its autonomy. Nonetheless, the privileging of unaccountable agents of policy transfer within policy transfer networks can lead to cultural imperialism through the back door via the transplantation of international policy agendas that are at odds with indigenous policy systems. A necessary criterion for the development of effective policy transfer networks must therefore be to ensure that such reforms are compatible with indigenous policy systems and settings.
The issue of policy compatibility is, of course, not the only potential obstacle to successful policy learning. This thesis stresses the importance of developing a rigorous implementation perspective in policy transfer analysis to ensure that policy transfer has taken place. This will allow for the identification of the type of learning organization conducive to the facilitation of successful lesson-drawing and the development of a model of prospective evaluation to guide effective lesson-drawing. Here this thesis will draw on the work of Evans (2012) who has identified three broad sets of variables or obstacles to transfer which he derives from the case study literature: “cognitive” obstacles in the predecision phase, “environmental” obstacles in the implementation phase and domestic and, increasingly, international public opinion. These variables interact in complex and often unexpected ways and inform the process of policy transfer. “Cognitive” obstacles refer to the process by which public policy problems are recognized and defined in the predecision phase, the rationality of the search conducted for ideas, the receptivity of existing policy actors, and systems to policy alternatives and the complexity of choosing an alternative. The most significant cognitive barriers for agents of policy change to overcome at this stage of policy development are the prevailing organizational culture and the need for effective cultural assimilation of policy alternatives.
“Environmental” obstacles refer to the absence of effective, cognitive, and elite mobilization strategies deployed by agents of policy change, the need for the development of cohesive policy networks to ensure successful policy-oriented learning, the broader structural constraints (institutional, political, economic and social) that impact and shape the process of lesson-drawing and the normal technical implementation constraints that constrain or facilitate the process of lesson-drawing.
The latter would include: coherent and consistent objectives; the incorporation of an adequate causal theory of policy development; the sensible allocation of financial resources; hierarchical integration within and among implementing organizations; clear decision rules underpinning the operation of implementing agencies; the recruitment of program officers with adequate skills/training; sufficient technical support; and the use of effective monitoring and evaluation systems including formal access by outsiders.
In Conclusion: Understanding Learning and Avoiding the “Learning Paradox”
In sum then, policy learning analysis describes, almost uniformly, a lateral ring of information “exchange” (Bennett 1991) that leads to policy change in a nonindigenous setting. The study of policy learning can only be distinctive from the analysis of normal forms of policy-making if its focuses on the movement of ideas between systems of governance in different countries. Policy learning is therefore best concerned with the study of features of contemporary policy change that are not otherwise explained. However, most accounts lack the methodological tools necessary to provide a rigorous understanding of the movement of ideas from one setting to another. Moreover, the understanding advanced on the process of policy learning are at best patchy and are vulnerable to the criticism that it is questionable that learning has actually occurred at all.
On the other hand, policy learning can be a rational and progressive learning activity but only if the program that is transferred is compatible with the value system of the recipient organization, based on the best possible evidence, culturally assimilated through comprehensive evaluation, and, builds on existing organizational strengths. It suggests, following Marsh and McConnell (2010), three dimensions of success – process, programmatic, and political – but with additional indicators that provide a more comprehensive understanding of the learning process. In particular, our understanding of process success is enriched through recognition of the importance of agents (policy transfer networks) in creating ideational structures and diffusing innovative practices and that issues of institutional design are crucial to the generation of successful policy outcomes. Moreover, our understanding of programmatic success is advanced through developing evaluative tools for understanding what can and cannot be transferred (policy compatibility), what the key obstacles to effective policy transfer may be to ensure implementation and in whose interest the program has been developed. And, finally our understanding of political success is progressed through recognition of the importance of elite and cognitive mobilization strategies in winning the war of ideas. In short, evidence is not a sufficient criterion for affecting policy change; politics matters and communicative strategies are central to the achievement of policy success.
- Bovens M, ‘t Hart P, Kuipers S (2006) The politics of policy evaluation. In: Moran M, Rein M, Goodin RE (eds) The Oxford handbook of public policy. Oxford University Press, Oxford, pp 319–335Google Scholar
- Carroll, P (2013) Policy transfer and accession: a comparison of three international governmental organisations. J Comp Policy Anal Res Pract :2–17Google Scholar
- Carroll P, Common R (eds) (2013) Policy transfer and learning in public policy and management. Routledge, OxfordGoogle Scholar
- Evans M, Marsh D (eds) (2012) Policy transfer and policy translation. Policy Studies Routledge/Taylor and FrancisGoogle Scholar
- Haas PM (ed) (1992) Knowledge, power and international policy coordination. Special issue of International OrganizationGoogle Scholar
- Hulsewé AFP (1979) China in Central Asia: the early stage 125 BC – AD 23: an annotated translation of chapters 61 and 96 of the history of the former Han dynasty. E. Brill, LeidenGoogle Scholar
- Ikenberry G (1990) The international spread of privatization policies: inducements, learning and policy “Bandwagoing”. Cited in Suleiman E, Waterbury J (eds) The political economy of public sector reform and privatization. Westview, Oxford, pp 88–110Google Scholar
- Ladi S (2002) Promoting change through research. Joseph Rowntree Foundation, YorkGoogle Scholar
- Ladi S (2004) Globalization, policy transfer and policy research institutes. Edward Elgar, CheltenhamGoogle Scholar
- Larmer B (2005) The center of the world. Foreign Policy 150:66–74Google Scholar
- Leftwich A (2006) Introducing politics. Polity Press. JackedflapGoogle Scholar
- Page EC (2000) Future governance and the literature on policy transfer and lesson drawing. Paper prepared for the ESRC future governance programme workshop on policy transfer, 28 JanuaryGoogle Scholar
- Rose R (2005) Learning from comparative public policy: a practical guide. Routledge, LondonGoogle Scholar
- Weyland KG (2007) Bounded rationality and policy diffusion. Princeton University Press, PrincetonGoogle Scholar
- Zhang YZ, Marsh D (2014) Learning by doing – the case of administrative reform. Policy Studies, Routledge, Taylor & Francis GroupGoogle Scholar