Tag Archives: United States

Policy Concepts in 1000 Words: Policy Transfer and Learning

(podcast download)

‘Policy learning’ describes the use of knowledge to inform policy decisions. That knowledge can be based on information regarding the current problem, lessons from the past or lessons from the experience of others. This is a political, not technical or objective, process (for example, see the ACF post). ‘Policy transfer’ describes the transfer of policy solutions or ideas from one place to another, such as by one government importing the policy in another country (note related terms such as ‘lesson-drawing’, ‘policy diffusion’ and ‘policy convergence’ – transfer is a catch-all, umbrella, term). Although these terms can be very closely related (one would hope that a government learns from the experiences of another before transferring policy) they can also operate relatively independently. For example, a government may decide not to transfer policy after learning from the experience of another, or it may transfer (or ‘emulate’) without really understanding why the exporting country had a successful experience (see the post on bounded rationality). Here are some major examples:

BOX 12.1

It is a topic that lends itself well to practical advice; the ‘how to’ of policymaking. For example, Richard Rose’s ‘practical guide’ explores 10 steps:

Rose 10 lessons rotated

The descriptive/ empirical side asks these sorts of questions:

From where are lessons drawn? In the US, the diffusion literature examines which states tend to innovate or emulate. Some countries are also known as innovators in certain fields – such as Sweden and the social democratic state, Germany on inflation control and the UK on privatization. The US (or its states) tends to be a major exporter of ideas. Some countries often learn consistently from the same source (such as the UK from the US). Studies tend to highlight the reasons for borrowing from certain countries – for example, they share an ideology, common problems or policy conditions. ‘Globalization’ has also reduced practical barriers to learning between countries.

Who is involved? Apart from the usual suspects (elected officials, civil servants, interest groups), we can identify the role of federal governments (for states), international organizations (for countries), ‘policy entrepreneurs’ (who use their experience in one country to sell that policy to another – such as the Harvard Business School professor travelling the world selling ‘new public management’), international networks of experts (who feed up ideas to their national governments), multinational corporations (who encourage the ‘race to the bottom’, or the reduction of taxes and regulations in many countries), and other countries (such as the US).

Why transfer? Is transfer voluntary? The Dolowitz/ Marsh continuum sums up the idea that some forms of transfer are more voluntary than others. ‘Lesson-drawing’ is about learning from another country’s experience without much pressure (see the book to explain why I scribbled out some of the text!). At the other end is coercion. They place ‘conditionality’ near that end of the spectrum, since the idea is that countries who are so desperate to borrow money from the International Monetary Fund will feel they have no choice but to accept the IMF’s conditions – which usually involves reducing the role/ size of the state (although note the difference between agreeing to those conditions and meeting them). ‘Obligated transfer’ is further to the left because, for example, member states sign up to be influenced by EU institutions. Indirect coercion describes countries who feel they have to follow the lead of others, simply to ‘keep up’ or to respond to the ‘externalities’ or ‘spillovers’ of the policies of the other country (they are often felt most by small countries which share a border with larger countries).

figure 12.1 DM continuum

What is transferred? How much is transferred? Transfer can range from the decision to completely duplicate the substantive aims and institutions associated with a major policy change, taking decades to complete, to the vague inspiration (or the very quick decision not to emulate and, instead, to learn ‘negative lessons’).  It can also be a cover for something you planned to do anyway – ‘international experience’ is a great selling point.

What determines the likelihood and success of policy transfer? For an importing government to be successful, it should study the exporting country’s policy – and political system – enough to know what made it a success and if that success is transferable. Often, this is not done (governments may emulate without being particularly diligent) or it is not possible, since the policy may only work under particular circumstances (and we may not always know what those circumstances are). Much also depends on the implementation of policy, particularly when the transfer is encouraged by one organization and accepted reluctantly by another (such as when the EU, with limited enforcement powers, puts pressure on recalcitrant member states).

These questions are best asked alongside the general questions we explore in policymaking studies, including:

  • Bounded rationality and Incrementalism – do governments engage in trial-and-error and learn from their own mistakes first?  Is learning and transfer restricted to the ‘most similar’ regions because there is no point in learning from countries radically different from our own?  Do some governments emulate without learning? Is transfer from another, more innovative, government a common rule of thumb?
  • Multi-level Governance – does the existence of more policymaking arenas produce more innovation and a greater demand for learning? Or, does the diffusion of power undermine the ability of a central government to adopt policies from others?
  • Punctuated equilibrium – is transfer a rare opportunity produced by the sudden and unpredictable attention to new ideas?

Further Reading:

I explore these issues (and Rose’s advice) in a paper examining what Japan can learn from the UK’s experience of regionalism. It includes a discussion (summarised from Keating et al – Paywall Green) of the extent to which policy converges in a devolved UK and how much of that we can attribute to transfer and/ or learning:

Keating et al 2012 summary from japan paper


Filed under 1000 words, agenda setting, Evidence Based Policymaking (EBPM), Japan, public policy, Scottish politics, UK politics and policy

Policy Concepts in 1000 Words: Bounded Rationality and Incrementalism

(podcast download)

A classic starting point in policy studies is to compare ideal-types (which might be ideals to aspire to) with the real world. The classic example is comprehensive (or synoptic) rationality. The idea is that elected policymakers translate their values into policy in a straightforward manner. They have a clear, coherent and rank-ordered set of policy preferences which neutral organizations carry out on their behalf. We can separate policymaker values from organizational facts. There are clear-cut and ordered stages to the process (aims are identified, the means to achieve those aims are produced and one is selected) and analysis of the policymaking context is comprehensive. This allows policymakers to maximize the benefits of policy to society in much the same way that an individual maximizes her own utility.

Its comparator is ‘bounded rationality’ (coined by Simon) which suggests that policymakers’ ability to make and implement decisions is more problematic. We question our ability to separate values and facts. We note that policymakers have multiple, often unclear, objectives which are difficult to rank in any meaningful way. We wonder if the policy process is so ordered and linear (or if policymakers sometimes select a solution that already exists to a problem defined for them). We know that policymaking organizations have limited knowledge and research capabilities; that they have to use major shortcuts to gather a limited amount of information in a limited time. We know not to seek policymaking perfection, but something that is good enough. We don’t ‘maximize’ – we ‘satisfice’.

We can use this discussion to go down two main paths. The first is empirical/ descriptive. Lindblom’s famous conclusion is that bounded rationality helps cause incrementalism[i]. Organizations use simplifying strategies, such as limiting policy analysis to a small number of policy choices which diverge incrementally from the status quo (based on the argument that it is better to analyse a few issues comprehensively than seek comprehensive coverage of all issues). They use trial and error. Rather than ranking preferences in advance, they test their willingness to trade off one aim for another when they make policy decisions. They use an incremental strategy as a rule of thumb: if a previous policy commanded widespread respect then policymakers recognise the costs (analytical and political) of a significant departure from it.

If we follow this empirical path, we want to know if policymaking, and its outcome, is incremental. This picture of policymaking seemed to be common wisdom for some time: Lindblom’s critics often bemoaned the problems with the outcome, not his analysis. We can also identify related terms such as path dependence, policy succession and inheritance before choice which, albeit in different ways, highlight the dependence of current policy decisions on those made in the past. More recently, punctuated equilibrium theory suggests that incrementalism is not the full story. Rather, we can identify a mix of ‘hyper-incrementalism’ and radical change; a huge number of small changes and a small number of huge changes. Studies of policy diffusion also suggest that bounded rationality often prompts governments to emulate the (often radical) policies of other governments without fully understanding their success. So, we can now identify, in the literature, a common focus on bounded rationality as a starting point, but some very different conclusions about policy change.

The second path is normative/ prescriptive. Lindblom’s early work was often criticised for selling incrementalism as good practice: focusing on radical options is futile if no-one will countenance them anyway; policy in a series of steps reduces serious mistakes; and, existing policy is good if based on wide agreement. It prompted intense debate, focused on the extent to which incrementalism was appropriate when governments had a mandate for, or a need to engage in, radical change. Lindblom’s thinking also changed to some extent, to reflect his diminishing belief that the US political system was pluralistic and therefore a vehicle for policy based on widespread agreement.

This debate took place when Lindblom was contrasting policymaking in the US with the Soviet Union. Incrementalism was partly an antidote to ‘central planning’. While that specific debate has aged, we can still identify a broader concern with the centralization of power. The ideal of comprehensive rationality includes an assumption that power is held centrally by policymakers whose decisions are carried out by neutral bureaucrats or other organizations. In other words, a central decision maker should control the policy process; power should reside in the hands of elites at the top/ the centre at the expense of other actors. This raises debates about the balance between central and local government power, particularly when both have electoral mandates. We should also consider our need to balance authority at the top with local knowledge at the bottom; to balance the delegation of policymaking to people who know best how to do it with the maintenance of a meaningful degree of accountability for the outcomes. For example, you can see this debate play out in studies which apply complexity theory to public policy. In practice, the descriptive/ prescriptive elements of these discussions become blurred. One may argue for particular arrangements not only based on principle but also because they appear to work.

Much of the postwar debate took place in the idiosyncratic US, but incrementalism (or, in some cases, inertia) has been identified as a defining feature in systems such as Japan, Italy and Germany. It is tempting to associate incrementalism with particular types of system, such as ‘consensus democracies’ or systems with diffused power and multiple veto points – in contrast to the ‘majoritarian’ or ‘top-down’ UK. Yet, incrementalism’s key features – bounded rationality; the necessity of bargaining and compromise between actors who have different information, different interests and conflicting views; and, the need to build on past policies – may be ‘universal’.

[i] Policymaking through non-radical steps. Note that the meaning of ‘incrementalism’ is not always clear and that Lindblom is not its only exponent – look out for Wildavsky too.


Filed under 1000 words, public policy, UK politics and policy

Policy Concepts in 1000 Words: Punctuated Equilibrium Theory

cloud punctuated equilibrium

See also What is Policy? and the Policy concepts in 1000 words series

(podcast download)

Policymaking can appear stable for long periods, only to be destabilised profoundly. Most policies can stay the same for long periods while a small number change quickly and dramatically. Or, policy change in one issue may be minimal for decades, followed by profound change which sets policy on an entirely new direction. The aim of Baumgartner and Jones’ punctuated equilibrium theory is to measure and explain these long periods of policymaking stability, and policy continuity, disrupted by short but intense periods of instability and change. The key concepts are:

Bounded rationality. Policymakers cannot consider all problems and their solutions at all times. For example, government ministers can only pay attention to a tiny proportion of the issues for which they are responsible. They ignore most and promote few to the top of their agenda.

Disproportionate attention. Policymakers often ignore issues or pay them an unusual amount of attention. The lack of attention to most issues helps explain why most policies may not change. Intense periods of attention to some issues may prompt new ways to understand and seek to solve old problems.

Power and agenda setting. Some groups try to maintain their privileged position by minimizing attention to the policy solutions which benefit them. Others seek to expand attention, to encourage new audiences and participants, to generate debate and new action.

Framing. Groups compete to influence how a problem is framed (understood, defined, categorized and measured) and therefore solved by policymakers. For example, it may be framed as a problem that has largely been solved, leaving the technical details of implementation to experts, or a crisis which should generate widespread attention and immediate action.

Policy monopolies. Groups may enjoy a ‘monopoly of understanding’ when policymakers accept their preferred way to frame an issue for long periods, perhaps even taking it for granted. This monopoly may be ‘institutionalised’ when rules are created and resources devoted to solving the policy problem on those terms.

Venue shopping. To challenge a monopoly in one venue (such as the executive, or one type of government at a particular level), groups may seek an audience in another (such as the legislature, the courts, or another type or level of government).

In Agendas and Instability (1993; 2009), Baumgartner and Jones, use a case study approach to examine these processes in detail. For example, in postwar US nuclear power, they identify a period of major public attention, focused on the pressing need to solve a policy problem, followed by minimal attention – for decades – when the problem appeared to be solved. The government inspired public enthusiasm for nuclear power as a solution to several problems – including the need to reduce energy bills, minimise dependence on other countries for oil, reduce air pollution, and boost employment and economic activity. This positive image, and general sense that the policy problem was solved, supported the formation of a post-war policy monopoly involving the experts implementing policy. Public, media and most government attention fell and the details of policy were left to certain (mainly private sector) experts, federal agencies and congressional committees. The monopoly was only challenged in the 1970s following environmental activist and scientific concern about nuclear safety. Groups used this new, negative, portrayal of the nuclear solution to generate concerned interest in new venues, including the courts, congressional committees and, particularly following a major accident at Three Mile Island, the public. The policy monopoly – the way in which nuclear power was framed, and the institutions established to implement policy – was destroyed.  Then, a new, negative, image became dominant for decades – and a post-war policy of power plant expansion was replaced by a moratorium and increased regulation. Only recently has the resurgence of nuclear power become a serious possibility.

In The Politics of Attention (2005), Jones and Baumgartner’s focus shifts to more general observations of selective attention (they highlight over ‘400,000 observations collected as part of the Policy Agendas Project’). Policymakers are unwilling to focus on certain issues for ideological and pragmatic reasons (e.g. some solutions may be too unpopular to consider; there is an established view within government about how to address the issue). They are also unable to pay attention because the focus on one issue means ignoring 99 others. Change may require a critical mass of attention to overcome the conservatism of decision makers and shift their attention from competing problems. If levels of external pressure reach this tipping point, they can cause major and infrequent punctuations rather than smaller and more regular policy changes: the burst in attention and communication becomes self-reinforcing; new approaches are considered; different ‘weights’ are applied to the same types of information; policy is driven ideologically by new actors; and/or the ‘new’ issue sparks off new conflicts between political actors. Information processing is therefore characterized by ‘stasis interrupted by bursts of innovation’ and policy responses are unpredictable and episodic rather than continuous. One key example is the annual budget process which, like many other examples of political activity, does not display a ‘normal distribution’ of cases. Instead, it is characterized by a huge number of minimal changes and a small number of huge changes.

The trend in Baumgartner and Jones’ work has been to move from the specific to the ‘universal’; from some cases in one country to many cases in many countries. Their initial assumption was that many of the processes they observed in case studies resulted from the peculiarities of the US system. Yet, concepts such as bounded rationality, selective attention, policy monopolies and venue shopping should be applicable to all political systems. Punctuated equilibrium theory helps us to balance a focus on the specific and the general. We can use these concepts to generate empirical questions about why the policymakers, institutions and venues of specific political systems prompt particular problems and solutions to be addressed and others to be ignored. We can also use them to identify the same overall patterns – based on a mixture of stasis, stability and continuity disrupted by innovation, instability, and change – in many systems.

To read more, click here to get a Green Access version of chapter 9 of this book discussed here.


Filed under 1000 words, agenda setting, public policy

Conceptual ‘brand-stretching’ and the Policy Agendas Project

Darren Halpin and I are working on a paper about ‘brand stretching’ (his phrase) in policy studies. It might begin like this:

“Baumgartner and Jones’ (1993) Agendas and Instability has become one of the most influential accounts of the policy process. They explain, in a convincing way, why stable political arrangements and policy continuity may be disrupted by often-sudden instability and potentially profound policy change.  The book provides a rich study, of policymaking over several decades, to illustrate these mechanisms at work. While it contains case studies in US politics, many of the concepts it uses, and processes it identifies (such as ‘venue shopping’ and ‘framing’) can be described as ‘universal’; applicable to many (if not most) cases in the US and other countries.  Indeed, Jones and Baumgartner’s (2005) Politics of Attention takes us in that direction by identifying the ‘general punctuation hypothesis’ and applying it to aggregate budget and legislative data. The US political system as a whole demonstrates these long periods of stasis and bursts of activity, producing a large number of minor policy changes and small number of major changes. Crucially, this quantitative exercise is theory driven and consistent with the original argument. It allows us to identify, systematically, these important policymaking patterns and explain why they occur. When driven by the original authors, the research agenda is clear and consistent and has stood the test of time.

However, the Policy Agendas Project has also become a multi-country, multi-author, comparative research program. Other authors examining over fifteen countries such as the US, UK, Denmark, Korea and Australia, are engaged in a comparative effort to understand the dynamics of policy change and stability.  This expansion presents two major challenges. First, international brand expansion needs go hand in hand with coordination. The PAP is one of many comparative theories which originated from US studies of the US. A key question for scholars studying the political systems of other countries is: do their insights travel well?  To answer the question, we need to know which aspects of US-based theories are derived from particular aspects of the US political system and which are ‘universal’ processes, concepts or causal factors broadly applicable to many systems.  For example, ‘bounded rationality’ (Simon, 1976) limits the extent to which all policymakers can pay attention to issues – so they must ignore most while promoting a small number to the top of their agenda – while the processes, institutions and histories of specific political systems help explain how and why some issues are addressed and others ignored.  The task for comparative policy scholars is to explore the extent to which we can separate, analytically, the universal from the specific elements and compare their effects on policy processes and outcomes.  This requires a common language which is precise and clear enough to help us produce separate studies and compare them in a meaningful way.

Second, however, that coordination and common language may not always exist. While Jones and Baumgartner often play an important part in the coordination and direction of country-level studies, they also play a less direct role in the gathering, interpretation and presentation of the data. New approaches and research questions have developed, and the project appears to have taken on a life of its own. On the one hand, this is a good development – more people with their own ideas produce the potential for theoretical and empirical innovation. On the other, it produces the potential for confusion, as people start to use the original terms in different ways and/ or use ‘policy agendas’ as an umbrella term under which to pursue their own pet projects (often through separate conferences, internal codebooks, and bespoke terminology and forms of analysis). We use the phrase ‘brand-stretching’ to describe instances where the attractiveness of an original idea is used to support scholarship that, in many respects, shows increasingly tenuous links with the original.

What characterises the comparative project is the large scale collection of quantitative data sets that map the way institutions and actors – including the media, legislatures, administrative systems, and the public – prioritise attention to policy issues. Armed with code books that systematise and categorise policy areas, scholars have coded numerous aspects of policy activity in many countries. From a methodological and data perspective, the project seems to stand in stark contrast with the originating idea. In particular, we identify a tendency for quantitative data to be gathered without the same theory-driven approach. New authors may identify interesting patterns of behaviour or outcomes without demonstrating the same understanding of the policy process and what causes these outcomes. They may introduce the research and data without even showing a working knowledge of the original punctuated equilibrium concepts. In this context, we identify the potential for new work in comparative politics to undermine existing work in policy studies.  For example, a comparative politics study may be set up on the assumption that the primary source of variation is by country rather than by policy area or issue.

Brand-stretching is not specific to the policy agendas project. Rather, it is a general unintended consequence of major theories ‘catching fire’ and (a) attracting the attention of a large group of scholars (b) in multiple countries, or (c) seeking to apply the same concept to many countries. For example, Poteete et al’s (2010) Working Together considers how to coordinate multiple research projects on the Institutional Analysis and Development framework (IAD), and Weible et al’s (2009) ‘Taking Stock’ considers how the Advocacy Coalition Framework (ACF) has changed after its application in over 80 studies.

We use this comparative theory-expansion experience to perform a similar ‘stock taking’ exercise for the comparative agendas project (CAP). First, we summarise the Baumgartner and Jones policy agendas project (PAP), identifying its main features: focus, concepts, methods, and findings. Second, we summarise the CAP and compare it to the original, considering the extent to which the original brand has been stretched. Third, we identify the main ways in which other theory/ project leaders have dealt with the potential for brand stretching – and apply these insights to the CAP”.

Hopefully it reads as a positive-but-concerned discussion – as a way to help protect the classics. Poteete et al provide a much more extensive discussion of this issue when discussing the management of the IAD.


Filed under agenda setting, public policy

Street Level Bureacrats

So, I went to a session at #apsa2013 on Lipsky and Street Level Bureaucrats (no, Lipsky wasn’t there) and it got me to thinking about a partial shift of thinking in this area. Consider my summary of Lipsky in Understanding Public Policy (p37):

Lipsky (1980) argues that policy is, to a large extent, made by the ‘street-level bureaucrats’ (including teachers, doctors, police officers, judges, and welfare officers) who deliver it. Bureaucrats are subject to an immense range of, often unclear, requirements laid down by regulations at the top, but are powerless to implement them all successfully (1980: 14). In other words, this is not necessarily an argument based on ‘disobedience’: committed workers do not have the resources to fulfil all of their job requirements (1980: xii). Instead, they use their discretion to establish routines to satisfy a proportion of central government objectives while preserving a sense of professional autonomy necessary to maintain morale. The irony is that the cumulative pressure associated with central government policy effectively provides implementers with a degree of freedom to manage their budgets and day-to-day activities. Therefore, policy change at the top will not necessarily translate to change at the bottom.

For me, the suggestion is that the routines are based on professional norms; each profession has its own formal and informal means to encourage professionals to act in a particular way. However, a lot of the discussion was about more modern studies of SLBs and a shift to a focus on street level organizations (SLOs). The argument here is that: (a) much of this work is now done by quasi or non-governmental organisations; (b) the rise of new public management has led to attempts to regulate that policy delivery through measures such as targets, contract management and performance management. It ties in neatly with the idea of a shift from government to governance which, in this case, refers to a shift from direct to indirect public service delivery. The phrase – from Evelyn Brodkin – that stuck with me is that performance management reduces policy down to a few aspects of the law. If I understood this correctly, it points to an important change in the Lipsky-style argument. When, under direct government delivery, the SLBs were the people reducing policy down into a smaller number of manageable acts, this role is often now performed by people higher up the chain when setting performance related targets. It reminded me of the ‘regulatory state’ style argument which speaks to the tension between professional discretion and rules-based action. This idea tied in neatly with another paper (Rubin/ Chiques) at the panel which suggested that there were fewer people in SLOs delivering policy with discretion. Rather, there were layers of management responsible for dealing with targets (although still, of course, there is an element of discretion when organisations game the system to meet targets). This informed the first paper (Schram et al) which talked about the routines evident in welfare policy work. Some of it was about familiar indicators of discretion to do with categorising people so that they didn’t lose out on social security payments. Some of it was about the routines associated simply with meeting targets and performing semi-automated tasks. So, Schram talked about a ‘dehumanising’ need to meet numerical targets and a form of discretion that was often ‘wide, not deep’. The papers were US based but the focus on a shift from direct delivery, by professions, to targets-based service delivery by quangos and/ or non-governmental organisations are relevant to countries like the UK.


Filed under public policy, UK politics and policy