The Limitations of ‘Policy Transfer’ and ‘Lesson Drawing’ for Public Policy Research

Abstract

The concepts of ‘lesson drawing’ and ‘policy transfer’ have become increasingly influential ways of understanding public policy, especially in the UK. However, the main proponents of the concepts, Rose for ‘lesson drawing’ and Dolowitz and Marsh for ‘policy transfer’, have difficulty in providing convincing answers to three questions that are important for them and those engaged in similar studies. First, can they be defined as distinctive forms of policy-making separate from other, more conventional, forms? ‘Lesson drawing’ is very similar to conventional accounts of ‘rational’ policy-making and ‘policy transfer’ is very difficult to define distinctly from many other forms of policy-making. Second, why does ‘lesson drawing’ and ‘policy transfer’ occur rather than some other form of policy-making? The proponents of ‘policy transfer’ put a set of diverse and conflicting theories under a common framework, obscuring differences between them. Third, what are the effects of ‘lesson drawing’ and ‘policy transfer’ on policy-making and how do they compare to other processes? Whilst the effect of more ‘lesson drawing’ seems to be more ‘rational’ policy-making, the effect of ‘policy transfer’ on policy ‘success’ and ‘failure’ is less clear. Dolowitz and Marsh redescribe aspects of ‘failure’ as different forms of ‘transfer’ rather than giving independent reasons for outcomes based on features of transfer processes. Overall, particularly in the case of ‘policy transfer’, researchers may be better off selecting from a range of alternative approaches than limiting themselves to these conceptual frameworks.

The concepts of ‘lesson drawing’ (Rose, 1991, 1993) and ‘policy transfer’ (Dolowitz and Marsh, 1996, 2000) have become increasingly influential ways of studying public policy, especially in the UK. They form an important part of the theoretical basis of the ESRC Future Governance Programme, a major initiative costing around £3.5m and encompassing 30 projects (ESRC, 2000; Page, 2000). The programme aims to draw lessons from policy initiatives and practices in different jurisdictions by looking at the potential for innovation by cross-national learning,¹ examines how policies work when they are adopted as the result of international treaties and obligations² and identifies processes and mechanisms of policy learning and transfer and how they work.³ Individual projects explore a range of issues and are not all similarly committed to ‘lesson drawing’ and ‘policy transfer’. However, most use the terms ‘lessons’, ‘learning’, ‘transfer’, ‘transferring’ in their title and refer to these concepts in their project summaries (ESRC, 2000). Policy transfer has been influential beyond the programme, forming the basis for other contemporary research looking at cross-jurisdictional influences on policy-making (Dolowitz et al., 2000; Radelli 2000; Stone 2000).

We suggest that the concepts of ‘lesson drawing’ (Rose, 1991, 1993) and especially ‘policy transfer’ (Dolowitz and Marsh, 1996, 2000), in their current forms, are of limited use for pursuing the aims of the programme and similar research. Our critique is not based on specifying alternative definitions to encompass empirical events or processes that the concepts in some way fail to recognise. Nor does it assert an alternative abstract definition or alternative explanatory model that is, in some way, a more interesting topic for research. Instead we suggest that, given the issues that researchers using ‘lesson drawing’ and ‘policy transfer’ are interested in exploring, the principal proponents of the concepts provide only partially convincing answers to three key questions. First, what are ‘lesson drawing’ and ‘policy transfer’ and how do they differ from other forms of policy-making? Second, why does ‘lesson drawing’ and ‘policy transfer’ occur rather than some other form of policy-making? Third, what are the effects of ‘lesson drawing’ and ‘policy transfer’ on policy-making and how do they compare to other processes? We address each question in turn.

What is ‘Lesson Drawing’ and ‘Policy Transfer’ and How Does it Differ from Other Forms of Policy-Making?

Researchers need to be able to identify distinctive processes and mechanisms that constitute ‘lesson drawing’ and ‘policy transfer’ to assess how much is happening in absolute terms and relative to other forms of search processes, decision-making and programme design constituting policy-making. A clear and distinctive definition is needed to evaluate claims about trends, for example that ‘policy transfer’ has increased in recent years (Dolowitz and Marsh, 1996, 2000). The value of the concepts is further enhanced if they offer a new perspective on policy-making, identifying phenomena or trends that have not previously been noted.

In Lesson Drawing in Public Policy, Rose explores how lessons are drawn through policy-makers' dissatisfaction with the status quo and decisions that a programme elsewhere may be capable of being put into effect in their environment (Rose, 1993, pp. 1–11, 21–3). A policy ‘lesson’ is defined as ‘more than a symbol invoked to sway opinion about a policy and more than a dependent variable telling a social scientist what is to be explained. A lesson is a detailed cause-and-effect description of a set of actions that government can consider in the light of experience elsewhere, including a prospective evaluation of whether what is done elsewhere could someday become effective here’ (Rose, 1993, p. 27). Several processes are involved in lesson drawing. ‘Copying’ involves ‘enacting a more or less intact program’ already in effect; adaptation is similar but involves ‘adjusting for contextual differences’; making a ‘hybrid’ consists of ‘combining elements of programs from two different places’; synthesis is ‘combining familiar elements from programs in a number of different places to create new programs’; and ‘inspiration’ is ‘using programs elsewhere as an intellectual stimulus to develop a novel program’ (Rose, 1993, p. 30). Rose claims that lesson drawing occurs across time and space and is both positive, leading to prescriptions about what ought to be done, and negative, in terms of what not to emulate (Rose, 1993, pp. ix–x).

‘Lesson drawing’ is very similar to conventional rational accounts of policy-making which stress that policy decisions are made about the pursuit of valued goals through structured interventions by public bodies or their agents. The decisions are based on searching for the means to pursue goals in a systematic and comprehensive manner, reviewing policy in the light of past experience and any other available information to make adjustments where necessary (see Hill, 1997, pp. 23–4; Howlett and Ramesh, 1995, pp. 140–41). Rose's analysis does not seem to extend much beyond this well-known framework, apart from suggesting that there are limits to how far policy-makers are able to carry out this activity in practice. The lack of a clear distinction makes finding evidence of ‘lesson drawing’, as opposed to rational policy-making in general, a difficult task. Rose argues that lesson drawing is widespread and seems to be increasing (Rose, 1993, pp. 3–4). But it is hard to think of any form of rational policy-making that does not, in some way, involve using knowledge about policies in another time or place to draw positive or negative lessons. Even rational policy-makers' preference for the status quo in their own jurisdiction could be seen as implicitly involving negative lessons about alternatives in other countries or in other times.

The ‘lesson drawing’ concept offers the potential for distinguishing rational policy-making from forms of apparently non-rational policy-making, where knowledge seems not to be used to pursue goals in a systematic way. Rational ‘lesson drawing’ provides a different conceptual approach to accounts which stress the organisational-cultural processes involved in learning, which often have more to do with rituals and legitimacy than with processes of optimisation (see Argyris and Schön, 1978, pp. 319–31). In this sense, the ‘lesson drawing’ perspective deepens concepts of rational policy-making and enables policy-makers' behaviour to be compared to a benchmark of ‘lesson drawing’ behaviour.

The definition given by proponents of ‘policy transfer’ is even broader, making it difficult to disentangle not only from ‘rational’ but also from a wide range of other concepts of policy-making. Dolowitz and Marsh draw on a range of authors who examine the spread of policies and learning processes (for other surveys, see Evans and Davies, 1999, and Stone, 1999). They state that ‘policy transfer, emulation and lesson drawing all refer to 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 and Marsh, 1996, p. 344). In more recent work, they place more emphasis on transfer between political systems than over time within a system, suggesting that ‘policy transfer [refers] to the process by which knowledge of policies, administrative arrangements, institutions and ideas in one political system (past or present) is used in the development of policies, administrative arrangements, institutions and ideas in another political system’ (Dolowitz, 2000, p. 3; Dolowitz and Marsh, 2000, p. 5). The processes of ‘policy transfer’ include ‘voluntary’ adoption, attempting to subsume Rose's ‘lesson drawing’, and ‘coercive transfer’, where a government or supranational institution encourages or even forces a government to adopt a policy (Dolowitz and Marsh, 1996, pp. 344–45). A very broad range of objects of transfer is suggested, including policy goals, structure and content, policy instruments or administrative techniques, institutions, ideology, ideas, attitudes and concepts and negative lessons (Dolowitz and Marsh, 1996, pp. 349–50). In later work this list is presented as ‘programmes, policies, institutions etc.’ (Dolowitz, 2000, p. 3).

‘Policy transfer’ attempts to subsume concepts about the alteration of domestic policy by external influences and ‘convergence’ of policy in different countries (Bennett, 1991, 1997). Such ideas of a country drawing on other jurisdictions' laws and experiences have been explored for a long time (see Jacoby, 2000, pp. 1–41). Well-known examples include Witte's account of railway regulation (1932), the development of public services (Barker, 1944), the transfer of institutions in post-colonial states (Lerner, 1964; Subramaniam, 1977) and British-US cross-national transfer of policies (Wolman, 1992). Furthermore, the spread of policies across space has been explored in the diffusion literature (Berry and Berry, 1999; Most and Starr, 1990), sociological organisation literature about the spread of similar forms of organisation across different bodies (Dimaggio and Powell, 1991, pp. 63–82) and the cybernetics literature about control in complex environments (see Beer, 1966, p. 289). The social learning literature resembles the ‘learning’ aspects of ‘policy transfer’, although these links are less explicitly acknowledged by Dolowitz and Marsh. Social learning has a number of variants (for a review, see Argyris and Schon, 1978, pp. 319–31). Hall (1993) has stressed levels and processes of ‘learning’ and their application, focusing on objects of learning in three different orders of changes: first, instrument-setting; second, policy technique; and third, paradigms or socially constructed policy goals. A literature on UK macro-economic policy learning has developed, influenced by Hall's account (Greener, 2001; Hay, 2001; Oliver, 1997; Oliver and Pemberton, 2001; Pemberton, 2000).

The proponents of ‘policy transfer’ claim that there has been an increase in its occurrence. They state that ‘the rapid growth in communications of all types since the Second World War has accelerated the process [of policy transfer]’ (Dolowitz and Marsh, 1996, p. 343). Whilst this statement could imply that the processes operating in cases of policy transfer occur more quickly now than before, Dolowitz and Marsh principally mean that there are more cases, arguing that ‘While there is no denying that the process of policy transfer is not new, it nonetheless appears that over the past decade or so, as technological advances have made it easier and faster for policy-makers to communicate with each other, the occurrences of policy transfer have increased’ (Dolowitz and Marsh, 2000, p. 6). Their second, related, claim is that the significance of policy transfer in policy-making has increased. Dolowitz argues that, ‘not surprisingly, the increase in the occurrence of policy transfer has led to an increased interest in the process. Certainly, as transfer is increasingly a feature of policy-making, it is important that the process be better understood’ (Dolowitz, 2000, p. 2). Similarly, Dolowitz and Marsh (2000, p. 5) state ‘Given that policy-makers appear to be increasingly relying upon policy transfer, it is something that anyone interested in, or studying, public policy needs to consider’.

To evaluate the first claim of increased ‘policy transfer’ over time, clear measures of transfer are needed to assess whether the amount of knowledge of other places utilised in policy-making has over time increased, decreased, remained the same or changed in some other way. Measures could then be used as part of a survey of ‘policy transfer’ in a policy sector or for a jurisdiction over a period of time. To evaluate the claim that ‘policy transfer’ is of increased significance some measure of the relative importance of ‘policy transfer’ compared to other processes in overall policy-making for a sector or jurisdiction over time is required. We suggest that distinctive measures cannot be derived from ‘policy transfer’ because the breadth of the concept makes it difficult to disentangle from many other processes of policy-making.

The difficulties of gathering evidence about ‘policy transfer’ are compounded by Dolowitz and Marsh's approach to selecting cases. They list a few cases of alleged ‘policy transfer’ (Dolowitz, 2000, pp. 1–2; Dolowitz and Marsh, 2000, pp. 6–7). Dolowitz suggests that ‘A few examples drawn almost at random [emphasis added] from the areas of public policy, political development and democratisation are illustrative’ (Dolowitz, 2000, p. 1). He notes examples from three ‘fields’ of policy-making. In the field he terms ‘public policy’, Dolowitz gives the example of UK social security welfare policy where ‘much of the Conservatives’ welfare-to-work program was transferred from the US‘; for ‘political development’ he notes cases where ‘as a condition for approving a loan, the International Monetary Fund often requires that a recipient country pursue particular economic policies’; in ‘democratisation’ he gives 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’ (Dolowitz, 2000, pp. 1–2). But these examples do not seem to be ‘drawn almost at random’ in the conventionally understood sense of selecting cases from a broader set using some form of probabilistic sampling technique. They are all positive instances where transfer is supposed to have happened at a particular time and do not constitute a convincing survey of policies in a sector or jurisdiction over time.

Why Does ‘Lesson Drawing’ and ‘Policy Transfer’ Occur and Why Does it Occur Rather than some Other Form of Policy-Making?

Proponents of ‘policy transfer’ see exploring the reasons for ‘transfer’ as a key part of their research agenda. The question of explanation appears less central to the concerns of Rose (1993, p. xi) who suggests ‘lesson drawing’ as a guide for policy-makers rather than entailing an explanation of why it occurs. To explain why the processes occur it is necessary to give reasons or causes that are separate from the processes themselves. However, the identification of what ‘lesson drawing’ and ‘policy transfer’ consist of is not fully separate from the reasons that they occur. The intention to learn from others' experience is both a reason for ‘lesson drawing’ and part of the process. Although there are limitations of this sort, both accounts specify additional causes or reasons to try and provide explanations for the processes.

Rose pursues a limited interest in explanation, setting out hypotheses about the likelihood of ‘lesson drawing’. They include the view that lesson drawing is more likely the less unique a programme is, the more substitutable the institutions needed to carry it out, the greater the equivalence of resources between governments (how similar their systems are), the simpler the cause and effect structure of a programme, the smaller the scale of change attempted, the greater the links in programmes between jurisdictions and the greater the congruity between the values of policy-makers and a programme's values (Rose 1993, pp. 118–42). The novelty of the insights offered by this framework is limited, being similar to the expectations of rational policy-making. However, the hypotheses do appear to be amenable for empirical investigation, even if researchers interested in this sort of issue may have to look elsewhere because of Rose's limited ambitions in explaining learning type processes.

In contrast, the proponents of ‘policy transfer’ see explanation as central to their concerns, suggesting that ‘policy transfer can be treated as a dependent or independent variable. A broad range of explanations are proposed and a number of factors are identified for the increase in ‘transfer’; advances in international communication seem especially important in their account (Dolowitz, 2000, p. 1; Dolowitz and Marsh, 2000, pp. 6–7). They present a ‘conceptual framework’ which gives reasons why ‘policy transfer’ occurs and suggests how different actors are involved (Dolowitz and Marsh, 2000, pp. 5–9). The framework, developed under the heading ‘The Dolowitz and Marsh Model’, presents a range of explanations for policy transfer. They present a ‘Why Transfer? Continuum’. At one end of the continuum is ‘Want to’ transfer involving ‘Voluntary’, ‘Lesson Drawing’ and ‘Perfect Rationality’. At the other end is ‘Have to’ transfer involving ‘Direct Imposition’. In between are ‘Mixtures’ consisting of ‘Lesson Drawing’ with ‘Bounded Rationality’ (Dolowitz and Marsh, 2000, p. 9). Later on in the same article they call the ‘continuum’ an ‘heuristic device that allows us to think more systematically about the processes involved. It helps for two related reasons. First, it identifies categories that can be used by researchers to frame their empirical work … Second, many cases of transfer involve both voluntary and coercive elements; the continuum helps us acknowledge that fact and this again deepens our knowledge of the process' (Dolowitz and Marsh, 2000, pp. 13–14).

The ‘policy transfer continuum’ collapses two conventional dimensions of difference between types of policy-making, one based on forms of rationality and the other based on degrees of coercion, into a single dimension of difference. In doing so, their approach obscures differences that the conventional public policy literature suggests are important (for surveys see Hill, 1997, pp. 98–101, and Howlett and Ramesh, 1995, p. 141). The first type of difference is between forms of rationality. Perfect rationality, involving full information about options in choice of strategies best to pursue goals, is conventionally contrasted with bounded rationality, in which decision-making is restricted by human and organisational characteristics resulting in efforts to find solutions that are good enough rather than fully optimal from a perfectly rational viewpoint. In the administrative context, this insight is usually associated with the work of Herbert Simon who argued that ‘Administrative theory is peculiarly 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, p. xxiv). Simon suggested that the factors bounding rationality are ‘psychological’, including incomplete knowledge of options and their consequences, and ‘organisational’, with organisational structures framing how choices are made (Simon, 1957, pp. 61–109).

In contrast, the difference between voluntary action and coercion is conventionally seen as being of a second type, to do with some form of political power, in the sense of one party's ability to get another party to do something by a range of methods including persuasion and direction (Hill, 1997, pp. 18–19). A possible overlap with the rationality dimension is political power exercised through the provision of information, but, with this caveat, the difference between voluntary action and coercion is normally seen as not being of the same kind as that between perfect and bounded rationality. Despite these conventions, Dolowitz and Marsh's framework associates ‘perfect rationality’ with ‘voluntary’, ‘lesson drawing’ and ‘transfer’ involving ‘mixtures’ of reasons containing ‘having to’ transfer elements as ‘bounded rationality’ (Dolowitz and Marsh, 2000, p. 9).

Collapsing the dimensions of difference on a single continuum means that the framework obscures the potential range of different explanations of the policy-making processes. It does not capture the differences between theories in which ‘transfer’ is ‘voluntary’, in the sense of not being forced by some international obligation in a treaty or external imposition, but involves ‘bounded rationality’ because it is heavily constrained by the sort of organisational limits identified by Simon. For example, Walker (1969) examined how prototypes serve as ‘rules of thumb’ or ‘analogies’ which help policy-makers simplify complex tasks confronting them as a way of explaining the form of the diffusion of policy amongst US states.

Researchers interested in conceptual, non-domestic or across-time influences on policy-making need not restrict themselves to using the ‘policy transfer’ framework. Alternatives are available from many sources, as examples from two contrasting literatures illustrate. The first type contains theories taking institutions as their core, the second type contains theories that focus on the power of ideas. We do not suggest that these theories are necessarily inconsistent with all aspects of the ‘policy transfer’ explanatory framework, rather that researchers are better off accessing them directly and that the ‘transfer’ framework obscures rather than illuminates differences between them.

There is considerable variety in institutional approaches, and many definitions of what counts as an institutional account (Hall and Taylor, 1996). However, defining these accounts broadly as involving some formal structure or less formal but stable pattern of behaviour, there are a number of approaches that offer well-developed explanations of how policy-making is mediated by institutions. Different schools of ‘institutionalism’ offer insights into the issue of who has power in the ‘coercive’ processes and why some people are recipients whilst others are the ones doing the coercing. Institutional analysis allows power relations within states and support or opposition for externally imposed policies to be examined. For example, the abolition of the Japanese ‘house’ system (‘ie’) as part of the policies during the occupation period after the Second World War challenged fundamental assumptions of Japanese society by potentially undermining the dominance of vertical over horizontal social relations. However, the persistence of administrative structures, such as official registries, ensured the social system still privileged family relationships and impacted on the evolution of the legal system (Wada, 2000). What might have been seen crudely as ‘coercive’ Westernisation and destruction of original traditions involved a process of creative adaptation, amending but not eliminating some traditional power structures (see also Badie, 2000).

Institutional analysis offers insights into how organisational structures affect learning processes. Levitt and March identify a range of learning processes for firms that are potentially applicable to a range of organisations (1998, pp. 321–23). Learning is affected by organisational characteristics and the relationships with other organisations. There is a range of possible ‘non-optimal’ processes, including superstitious learning encouraged by ritualistic behaviour and competency traps where path dependencies prevent the search for a fuller set of alternative policy options (Levitt and March, 1998, p. 325; see also Hall and Taylor, 1996, pp. 946–47). In a similar vein, Knill (1999) suggests administrative structure in Germany offers different potential for learning than the ‘amateurist’ administrative tradition of Britain, and Hood (1996) notes how the British practice of rotating civil servants across departments and domains promotes internal ‘learning processes’ within the central state.

An alternative literature examines the ‘power of ideas’ and policy-making (for a critique, see Blyth, 1997). Accounts of the spread of ideas often combine ideas-based and institutional accounts by invoking the networks of actors involved in learning and transfer (Bennett, 1991, p. 224). It is argued that the nature of the network is important in affecting how learning occurs and ideas spread. Networks with higher connectivity are likely to promote more learning than those with fewer links. Examples of these sorts of theory include the ‘advocacy coalitions’ and ‘epistemic communities’ approaches that stress the ways in which ‘policy learning’ and ‘policy expertise’ are structured (Sabatier, 1988; Haas, 1992). Such analysis involves mapping actors, analysing how these coalitions update their knowledge and illustrating their influence over policy formation and implementation.

The literatures on various forms of globalisation and internationalisation processes draw on both ideas- and institution-based accounts to offer various explanations for how policy-makers are influenced by forces beyond the domestic context. Globalisation is said to be establishing new economic, political and social circumstances that change state powers and the context in which the state operates (Rosenau, 1990; Mann, 1997). Some theories regard governments' learning from competitor states as the result of international locational competition and decreased national policy autonomy (Hoberg, 1991). Competitive adjustment seems to be a form of coercion but is not the same as direction from above. In some accounts, the state is seen as primarily attempting to boost international competitiveness within an open, global, competitive economy. These transnational forces compel states to imitate successful competitors (Cerny, 1997).

Similarly, the expanding literature on ‘Europeanisation’ has shown a particular concern with the way in which EU processes impact on domestic policy-making, including the administrative fusion of officials whose EU and national interests become interwoven (Wessels, 1997), the rebalancing of domestic coalitions in the face of additional or reduced resources because of EU processes (Knill and Lehmkuhl, 2002) and an emerging EU governing mode of the ‘open method of co-ordination’ (Hodson and Maher, 2001). Policy sector changes involving these processes include the liberalisation of network industries, changes in competition law and regionalisation (Héritier, 1997; Börzel, 1999; Schmidt, 1998; Lodge, 2000; Eyre and Lodge, 2000; Thatcher, 2000).

What Are the Effects of ‘Lesson Drawing’ and ‘Policy Transfer’ on Policy-Making Compared to Other Processes?

Researchers need to be able to distinguish the effects on policy from their ‘lesson drawing’ or ‘policy transfer’ causes and to disentangle their effects from the effects of other types of policy-making. Furthermore, Rose and some others involved in the Future Governance Programme are interested in ‘lesson drawing’ as a normative project to improve policy-making (Rose, 1993, pp. 1–18; Page, 2000). The agenda is similar to that of ‘evidence based’ policy-making which attempts to improve policy, in part, by getting the results of research on the effects of different policies incorporated in decision-making (Cabinet Office, 1999; Davies, Nutley and Smith, 2000). Such initiatives present the risk of being little more than exhortations that policy-makers should take all relevant information from their own experience and the experience of others into account in reaching a policy decision. However, defining Rose's perspective more narrowly and distinctively as a ‘how to’ guide could involve identifying techniques to improve ‘lesson drawing’, such as linking different sources of data or developing search tools to help overcome information overload. It appears that current research may be moving in this direction (ESRC, 2000).

Proponents of ‘policy transfer’, whilst acknowledging that it is not the sole explanation of any or most policy, argue that it has high contemporary salience and that, ‘As such, when we are analysing policy change we always need to ask the question: Is policy transfer involved?’ (Dolowitz and Marsh, 2000, p. 21). In particular, they suggest that ‘transfer may shape policy change’ and may ‘also lead to implementation failure’ (Dolowitz and Marsh, 2000, p. 21). A key part of their analysis examines the relationship between ‘policy transfer’ and policy ‘success’ and ‘failure’. After acknowledging the considerable problems of identifying what constitutes a success or failure, they state, ‘at this stage, we shall restrict ourselves to concentrating upon the extent to which policy transfer achieves the aims set by a government when they engaged in transfer, or is perceived as a success by the key actors involved in the policy area’ (Dolowitz and Marsh, 2000, p. 17).

Three factors contribute to ‘policy failure’ in their analysis (Dolowitz and Marsh, 2000, p. 17). First, in the case of ‘uninformed transfer’, the borrowing country may have insufficient information about the policy/institution and how it operates in the country from where it is being/was transferred. Second, in the case of ‘incomplete transfer’, ‘crucial elements’ of what made the policy or institutional structure a success in the originating country may not have been transferred, leading to ‘failure’. Third, in the case of ‘inappropriate transfer’, insufficient attention regarding social, economic, political and ideological differences between the ‘transferring’ and the ‘borrowing’ country leads to ‘failure’.

An overview of Dolowitz and Marsh's arguments about the link between ‘policy transfer’ and ‘policy failures’, illustrated by their chosen case of the British Child Support Agency (CSA) and ‘transfer’ from the US and elsewhere, is given in Table 1, along with a list of the problems inherent in their analysis. Their account does not develop an explanation of ‘failure’ as brought about by features of the process of ‘policy transfer’, for example in terms of how institutional features of the British state created bounded rationality in learning processes. Instead they redescribe aspects of policy ‘failure’ as ‘uninformed’ and ‘incomplete’ types of ‘policy transfer’. They introduce explanations of ‘failure’ which go beyond ‘policy transfer’, such as British policy-makers' revenue-raising goals, which led the agency to pursue ‘easy’ targets, reducing the legitimacy of the system in the eyes of the public. This last feature of policy ‘failure’ would be more straightforwardly understood as the British government having a goal that led it to develop the CSA in a way that proved to be unpopular rather than dressing the explanation in the language of ‘policy transfer’.

Alternative accounts exploring the consequences of policy-making processes for policy outcomes focus more directly on how processes influence outcomes. Institutional accounts based on ‘path dependence’ suggest that policy changes in general face substantial difficulties because of fixed costs, resource-dependent constituencies and established standard operating procedures (North, 1990; Pierson, 2000). The literature on organisation theory has warned of pathological forms of learning, such as overly backward and conservative learning or, in contrast, overly missionary and forward-looking learning (Deutsch, 1963). Some accounts have attempted to improve processes in order to allow organisations to process double-loop learning to reflect on their ‘theory-in-use’ (Argyris and Schön, 1978, p. 22). The notion of a ‘policy irritant’, as suggested by Teubner (1998), encapsulates the complexity of introducing policy and law into established institutional contexts. It stresses the importance of the transforming character of the ‘transferred policy’ on the broader context into which the policy is being ‘transferred’ and cross-national contextual differences. For example, differences between coordinated and uncoordinated market economies are likely to make the systems respond differently to the adoption of a similar economic model in different countries (Soskice, 1999).

Table 1

‘Policy Transfer’ as an Explanation of Policy ‘Failure’: The Case of the British Child Support Agency

Type of transfer	Dolowitz and Marsh argument	Critique
Uninformed transfer	Sole focus on Wisconsin example led to lack of attention on other US states and the Australian equivalent (pp. 18–19)	Does not fully develop an explanation of ‘policy failure’ in terms of specific features of the process of ‘transfer’. For example, institutional features of the British state may have led to bounded rationality in learning processes
Incomplete transfer	In the British CSA, administrative systems replaced the courts that were used in the US case, eliminating an ‘escape valve’ to advance compliance and decrease dissatisfaction with CSA decisions (p. 19)	Does not explain policy failure in terms of the process of ‘transfer’ but redescribes the ‘failure’ as a form of ‘incomplete transfer’
Inappropriate transfer	British bias towards reducing the Public Sector Borrowing Requirement led the pursuit of revenue collection to compromise the CSA's other goals, leading to implementation problems (p. 20)	Does not explain policy failure in terms of the process of ‘transfer’ but accounts for failure through the aims of policy-makers in Britain which reduced the CSA's legitimacy in the eyes of the public

Conclusion

The concepts of ‘lesson drawing’ and ‘policy transfer’ are difficult to define distinctly from other forms of policy-making. ‘Lesson drawing’ is broadly similar to conventional accounts of how policy-makers strive to make ‘rational’ policy decisions, especially if ‘lesson drawing’ is defined broadly as drawing positive and negative lessons across time and space (Rose, 1993, pp. ix–x). Whilst ‘lesson drawing’ offers the potential for comparing how policy-makers behave with the expectations of ‘lesson drawing’, ‘policy transfer’ is even broader in scope, encapsulating ‘lessons’, other forms of ‘voluntary’ adoption and ‘coercive’ processes. ‘Policy transfer’ is very difficult to disentangle from other forms of policy-making and researchers will find it very difficult to form clear measures of practical use for assessing claims about changes in the importance of ‘transfer’.

Focusing on certain elements of the ‘policy transfer’ concept and developing clearer measures of ‘transfer’ might be one way to develop the approach. Evans and Davies (1999, p. 382) have proposed that effort should be directed to validating whether ‘transfer’ has occurred and assessing the extent of ‘non-transfer’. Bennett (1997, pp. 214–15) develops an account of how to substantiate transfer by evaluating whether it can be demonstrated that idiosyncratic domestic factors are not independently responsible for policy adoption, that similar adoption is not the result of cross-national forces with separate effects in different states, that policy-makers' are aware of policies in other areas and that evidence from elsewhere is utilised within the domestic policy debate. The Future Governance Programme Director has proposed narrowing ‘transfer’ and ‘learning’ perspective down on the transposition of ‘policies’ and ‘practices’ already in operation in one system to another, rather than ‘ideas’ or ‘knowledge’ (Page, 2000). A similar approach is adopted by James (2001) in exploring the extent to which the ‘executive agency’ model of public service delivery, as developed in the UK, was emulated in different countries.

Rose is less ambitious in offering explanations of ‘lesson drawing’ than proponents of ‘policy transfer’ who put a set of diverse and conflicting theories under a common framework. Rather than just a case of old wine in new bottles, the framework is an unpalatable cocktail of different types of beverage. The framework obscures differences between theories and might lead researchers who follow it to neglect the variety of theories that are available. In this sense, ‘policy transfer’ is less than the sum of its parts.

The effect of more ‘lesson drawing’ seems to be that policy-making works more like the systems conventionally described in rational accounts. The production of ‘how to’ guides specifying mechanisms for improving ‘learning’, including sources and ways to analyse evidence, would appear to be a way researchers contribute to its fulfilment. However, the use of ‘policy transfer’ to explain ‘policy change’ and policy ‘success’ or ‘failure’ does not adequately separate the policy ‘success’ or ‘failure’ being explained from processes of ‘policy transfer’. Instead, Dolowitz and Marsh redescribe aspects of ‘failure’ as some form of ‘transfer’. At other points, they evoke explanations that go beyond features of the process of ‘transfer’ to include the aims of policy-makers. Researchers may be better off using alternative theories focusing more directly on the effects of learning processes or styles of policy-making on policy outcomes.

Footnotes

1

Projects examine taxation, housing, social insurance, prisons, performance measurement, fisheries management, environmental protection, conflict resolution and improving standards in public life.

2

Projects examine European regional and social policy, money laundering, financial regulation, education, housing and public health in developing countries.

3

Projects examine contractors for information communication technologies, law and order policies, workfare policies, European utilities and regulation policies, public management structures and higher education structures.

References

Argyris

Schön

(1978) Organizational Learning: a Theory of Action Perspective. Reading MA: Addison Wesley.

Badie

(2000) The Imported State. Stanford CA: Stanford University Press.

Barker

(1944) The Development of Public Services in Western Europe. London: Oxford University Press.

Beer

(1966) Decision and Control. London: Wiley.

Bennett

C. J.

(1991) ‘Review Article: What is Policy Convergence and What Causes It?’, British Journal of Political Science, 21, 215–33.

Bennett

C. J.

(1997) ‘Understanding Ripple Effects: the Cross-National Adoption of Policy Instruments for Bureaucratic Accountability’, Governance, 10 (3), 213–33.

Berry

F. C.

Berry

W. D.

(1999) ‘Innovation and Diffusion Models in Policy Research’, in Sabatier

P. A.

(ed.) Theories of the Policy Process. Boulder CO: Westview.

Blyth

(1997) ‘Any More Bright Ideas? The Ideational Turn of Comparative Political Economy’, Comparative Politics, 29 (2), 229–50.

Börzel

(1999) ‘Towards Convergence in Europe? Institutional Adaptation to Europeanization in Germany and Spain’, Journal of Common Market Studies, 37 (4), 573–96.

10.

Cabinet Office (1999) Modernising Government, Cm 4310. London: Stationary Office.

11.

Cerny

(1997) ‘Paradoxes of the Competition State: the Dynamics of Political Globalization’, Government and Opposition, 32 (2), 251–74.

12.

Davies

H. T. O.

Nutley

S. M.

Smith

P. C.

(eds) (2000) What Works? Evidence Based Policy and Practice in Public Services. Bristol: Polity.

13.

Deutsch

(1963) The Nerves of Government: Models of Political Communication and Control. New York: Free Press.

14.

DiMaggio

P. J.

Powell

W. W.

(1991) ‘The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organisational Fields’, in Powel

W. W.

DiMaggio

P. J.

, The New Institutionalism in Organisational Analysis. Chicago IL: University of Chicago Press, pp. 63–82.

15.

Dolowitz

(2000) ‘Introduction’, Governance, 13 (1), 1–4.

16.

Dolowitz

Marsh

(2000) ‘Learning from Abroad: the Role of Policy Transfer in Contemporary Policy-Making’, Governance, 13 (1), 5–24.

17.

Dolowitz

Marsh

(1996) ‘Who Learns What from Whom: a Review of the Policy Transfer Literature’, Political Studies, 44, 343–57.

18.

Dolowitz

D. P.

Hulme

Nellis

O'Neal

(2000) Policy Transfer and British Social Policy. Buckingham: Open University Press.

19.

ESRC (2000) Introduction to the Future Governance Research Programme, http://www.hull.ac.uk/futgov/BasicNavigation/Index.html.

20.

Evans

Davies

(1999) ‘Understanding Policy Transfer: a Multi-Level, Multi-Disciplinary Perspective’, Public Administration, 77 (2), 361–85.

21.

Eyre

Lodge

(2000) ‘National Tunes and European Melody – Competition Law Reform in the UK and Germany’, Journal of European Public Policy, 7 (1), 63–75.

22.

Greener

(2001) ‘Social Learning and Macroeconomic Policy in Britain’, Journal of Public Policy, 21 (2), 1–22.

23.

Haas

(1992) ‘Introduction: Epistemic Communities and International Policy Coordination’, International Organization, 46 (1), 1–35.

24.

Hall

P. A.

(1993) ‘Policy Paradigms, Social Learning and the State – the Case of Economic Policy-making’, Comparative Politics, 25 (3), 275–96.

25.

Hall

P. A.

Taylor

R. A.

(1996) ‘Political Science and the Three Institutionalisms’, Political Studies, 44 (5), 936–57.

26.

Hay

(2001) ‘The “Crisis” of Keynesianism and the Rise of Neoliberalism in Britain: an Ideational Institutionalist Approach’, in Campbell

J. L.

Pedersen

O. K.

(eds) The Rise of Neoliberalism and Institutional Analysis. Princeton NJ: Princeton University Press.

27.

Héritier

(1997) ‘Market-making Policy in Europe: Its Impact on Member State Policies. The Case of Road Haulage in Britain, the Netherlands, Germany and Italy’, Journal of European Public Policy, 4 (4), 539–55.

28.

Hill

(1997) The Policy Process in the Modern State. Third edition. London: Prentice Hall/Harvester Wheatsheaf.

29.

Hoberg

(1991) ‘Sleeping with an Elephant: the American Influence on Canadian Environmental Regulation’, Journal of Public Policy, 11 (1), 107–32.

30.

Hodson

Maher

(2001) ‘The Open Method as a New Mode of Governance: the Case of Soft Economic Policy Co-ordination’, Journal of Common Market Studies, 39 (4), 719–46.

31.

Hood

(1996) ‘United Kingdom: from Second Chance to Near-Miss Learning’, in Olsen

J. P.

Peters

G. B.

(eds) Lessons from Experience. Oslo: Scandinavian University Press.

32.

Howlett

Ramesh

(1995) Studying Public Policy: Policy Cycles and Policy Subsystems. Oxford: Oxford University Press.

33.

Jacoby

(2000) Imitation and Politics. Ithaca NY: Cornell University Press.

34.

James

(2000) ‘Regulation Inside Government: Public Interest Justifications and Regulatory Failures’, Public Administration, 78 (2), 327–43.

35.

James

(2001) ‘Business Models and the Transfer of Business-like Central Government Agencies’, Governance, 14, 2.

36.

Knill

(1999) ‘Explaining Cross-National Variance in Administrative Reform: Autonomous Versus Instrumental Bureaucracies’, Journal of Public Policy, 19 (2), 113–39.

37.

Knill

(2001) The Europeanisation of National Administrations. Cambridge: Cambridge University Press.

38.

Knill

Lehmkuhl

(2002) ‘The National Impact of European Regulatory Policy: Three Europeanisation Mechanisms’, European Journal of Political Research, 41, 255–80.

39.

Levitt

March

(1998) ‘Organizational Learning’, Annual Review of Sociology, 14, 319–40.

40.

Lerner

(1964) ‘The Transfer of Institutions’, in Hamilton

(ed.) The Transfer of Institutions. Durham NC: Duke University Press.

41.

Lodge

(2000) ‘Isomorphism of National Policies? The ‘Europeanisation’ of German Competition and Public Procurement Law’, West European Politics, 23 (1), 89–107.

42.

Mann

(1997) ‘Has Globalization Ended the Rise and Rise of the Nation-State’, Review of International Political Economy, 4 (3), 472–96.

43.

Most

B. A.

Starr

(1990) ‘Theoretical and Logical Issues in the Study of International Diffusion’, Journal of Theoretical Politics, 2 (4), 391–412.

44.

North

(1990) Institutions, Institutional Change and Economic Performance. Cambridge: Cambridge University Press.

45.

Oliver

(1997) Whatever Happened to Monetarism? Aldershot: Ashgate.

46.

Oliver

Pemberton

(2001) ‘Learning and Change in Twentieth Century British Economic Policy’, Paper presented to the European Association for Evolutionary Political Economy Conference, Siena, 8–11 November 2001.

47.

Page

E. C.

(2000) ‘Future Governance and the Literature on Policy Transfer and Lesson Drawing’, paper to the ESRC Future Governance Programme Workshop, London, 28 January 2000. www.futuregovernance.ac.uk.

48.

Pemberton

(2000) ‘Policy Networks and Policy Learning: UK Economic Policy in the 1960s and 1970s’, Public Administration, 78 (4), 771–92.

49.

Pierson

(2000) ‘Increasing Returns, Path Dependence, and the Study of Politics’, American Political Science Review, 94 (2), 251–67.

50.

Rose

(1991) ‘What is Lesson Drawing?’, Journal of Public Policy, 11 (1), 3–30.

51.

Rose

(1993) Lesson-Drawing in Public Policy. Chatham NJ: Chatham House.

52.

Rosenau

J. N.

(1990) Turbulence in World Politics: a Theory of Change and Continuity. Princeton NJ: Princeton University Press.

53.

Sabatier

P. A.

(1988) ‘An Advocacy Coalition Framework of Policy Change and the Role of Policy Learning Therein’, Policy Sciences, 21, 128–68.

54.

Schmidt

(1998) Liberalisierung in Europa. Frankfurt/M: Campus.

55.

Simon

(1957) Administrative Behaviour. Fourth edition. New York: Basic Books.

56.

Soskice

(1999) ‘Divergent Production Regimes: Co-ordinated and Unco-ordinated Market Economies in the 1980s and 1990s’, in Kitschelt

Lange

Marks

Stephens

J. D.

(eds) Continuity and Change in Contemporary Capitalism. Cambridge: Cambridge University Press.

57.

Stone

(2000) ‘Non-Governmental Policy Transfer: the Strategies of Independent Policy Institutes’, Governance, 13 (1), 45–62.

58.

Stone

(1999) ‘Lesson Drawing and Policy Transfer’, Politics, 19 (1), 51–59.

59.

Subramaniam

(1977) Transplanted Indo-British Administration. New Delhi: Ashish.

60.

Teubner

(1998) ‘Legal Irritants: Good Faith in British Law or How Unifying Law Ends Up In New Divergences’, Modern Law Review, 61 (1), 11–32.

61.

Teubner

(1987) ‘Juridification: Concepts, Aspects, Limits and Solutions’, in Teubner

(ed.). Juridification of Social Spheres. Berlin: Walter de Gruyter.

62.

Thatcher

(2000) The Politics of Telecommunications. Oxford: Oxford University Press.

63.

Wada

(2000) ‘Die Abschaffung der Institution des Hauses (ie) unter den Besatzungsmächten oder Das Koseki, ein Januskopf’, Recht in Japan, Heft 12, Nomos.

64.

Walker

(1969) ‘The Diffusion of Innovations among the American States’, American Political Science Review, 63, 880–99.

65.

Wessels

(1997) ‘An Ever Closer Fusion? A Dynamic Macropolitical View of the Integration Process’, Journal of Common Market Studies, 35 (2), 267–99.

66.

Witte

(1932) Eisenbahn und Staat – Ein Vergleich der Europaschen und Nordamerikanischen Eisenbahnorganisationen in ihrem Verhaltnis zum Staat. Jena: Gustav Fischer.

67.

Wolman

(1992) ‘Understanding Cross-National Policy Transfers: the Case of Britain and the US’, Governance, 5 (1), 27–45.