Policy Transfer, Policy Learning, Policy Convergence, Policy Diffusion
Summary = 1 page, full document = 15
Summary:
Refers to the evidence for - and causes of - similarities in policy across regions.
A variety of terms are in use – transfer is used as an umbrella term for the rest.
Learning may be voluntary, convergence and diffusion may suggest a relative
absence of trans-national forces.
The actors involved are a combination of usual suspects (elected officials, parties,
civil service, etc.) as well as “policy entrepreneurs” and supra-national
organisations.
The nature of transfer can be seen on a continuum, from voluntary to coercive.
This includes indirect coercion which may suggest a perceived need to transfer,
often without exerted influence.
A variety of policy-related factors may be transferred – from ideologies and
wholesale programmes to administrative arrangements or just broad ideas.
Negative lessons may also be learned.
The level of transfer ranges from complete duplication to broad inspiration.
Policy conditions, geography and ideology affect the willingness to study other
regions.
Transfer is most likely if the policy is simple, the values of borrower and lender
coincide and the political structures/ administrative arrangements are similar.
There is a clear link between the transfer literature and issues of rationality,
governance and agenda-setting.
This lecture considers global policy convergence and the extent to which policy change
can be explained by a rational process of policy learning across states. It covers the
spectrum of policy transfer (from voluntary to coercive), the assumption of rationality
and the importance of country level differences. A good starting point (which these notes
draw on heavily for structure) is Dolowitz and Marsh (1996 and 2000), although the list
of references at the end should show you that there is no shortage of writing on the
subject. Most are available through Google Scholar.
But What Is It?
We have at least 4 classifications of policy change here:
1. Policy Convergence – perhaps the broadest of the 4 terms. Bennet’s (1991)
starting point is the comparative politics literature which identifies similarities
across a number of countries in terms of policy goals, content, instruments,
outcomes and/ or styles. There is the potential for convergence to refer to policy
change which occurs relatively independently of other countries. In other words,
countries with similar policy problems could address them in similar ways
without there being policy transfer. For example, Hoberg (2001: 127) lists
“parallel domestic pressures” and Holzinger and Knill (2005: 780) “independent
problem solving” as reasons for convergence, while Stone (1999: 52) suggests
that convergence and diffusion both imply a passive role for elites, with structural
factors more important.
Bennett (1991: 231) suggests that while we shouldn’t assume a trans-national
explanation we should not call the phenomenon “convergence” which relates to a
direct trans-national link. Bennet (1991: 217-220) suggests that the literature
identifies 4 main causes of convergence – emulation, policy networks with
international connections, harmonisation (often facilitated by supra-national
organisations such as the EU) and penetration (coercion): “it is not enough to say
that comparable conditions produce comparable problems which produce
comparable policies” and convergence suggests more than similarity (it suggests
movement towards similarity).
2. Policy Learning – this is the focus of Rose (1993) who discusses lesson-drawing
as a source of policy change. While Rose points out that lessons can be drawn
across time (i.e. from regioni A’s past experience)ii the focus is on learning across
space (i.e. from other regions). For Rose (1993: 30) the extent of learning varies,
from complete duplication, to adaptation (taking different starting points into
account), making a hybrid from the exporting and importing region, extracting
some aspects of an exporting programme and applying them internally, and just
using lessons from other regions as a broad inspiration from policy change. We
can also assume a level of “negative lesson-drawing” in which lessons learned
about region B dissuade policy makers in region A from taking a similar course of
action (Stone, 1999: 52 suggests lessons were learned from the UK about how not
to address BSE).
3. Policy Transfer – see the over-arching definition by Dolowitz and Marsh (2000:
5; see also 1996: 344): “the process by which knowledge about 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 and Marsh explore
the explanations of transfer, including the level of coercion, the actors involved,
the thing that is transferred, the source, the degree of transfer and so on.
4. Policy Diffusion – from what I can see, this term is used most often when
referring to (in a sense) internal policy transfer. For example, it has been used to
refer to a gradual process of policy convergence between states within the US and
is associated strongly with Walker (1969). Maybe the imagery resembles
dropping a small amount of ink into a large glass of water (?). Bennett (1991:
220) suggests that diffusion refers to similar adoptions of policy without evidence
of emulation.
In all 4 cases it does not do to get too involved in the semantics of the use of the term. As
you can see from the definition of transfer, the danger is that we get too bogged down in
trying to provide an exhaustive definition that we lose years of our lives and lose out on a
concise and catchy definition. It is enough to know that a precise definition is elusive,
with greater coverage at the expense of clarity (as with most definitions). We will try to
avoid this problem and stay focussed on the bigger picture - stability and change. Be
comforted by Dolowitz and Marsh (1996: 344) who read all of the literature for you and
still conclude “we do not think that the difference in nomenclature is overly significant”
(with the exception of policy learning which perhaps suggests voluntary rather than
coercive transfer).
Who Does It?
While a clear distinction is impossible, it may be worth highlighting those involved in the
export or selling of ideas as opposed to those importing. This may be particularly
important when we discuss the nature of transfer – i.e. is it voluntary or coercive?
Dolowitz and Marsh (2000: 10) suggest that there tend to be clear borrowers and lenders
but that this is not a hard and fast rule. Countries like Canada and the UK who tend to
borrow from the US also “lend” some polices. We should also see that e.g. a country like
New Zealand can initially borrow aspects of new public management but then become a
beacon for subsequent lending.
As Dolowitz and Marsh (1996: 345) suggest, many of the actors involved could be
associated generally with the domestic policy process. Thus, “elected officials, political
parties, bureaucrats/ civil servants, pressure groups” are obviously involved because these
are key players in the process regardless of our lecture topic. This leaves us with two
relatively unusual characters:
1. Policy entrepreneurs/ experts – this can relate to prominent individuals/
consultantsiii, think tanks or policy networks which span more than one region.
Dolowitz and Marsh (2000: 10) stray into prescriptive mode when discussing the
role of consultants who often sell “best practice” without regard to the political
circumstances of the importing country. They also point to the increasing role for
NGOs (non-governmental organisations) in international policy discourse.
Bennett (1991: 224) highlights the role of international policy communities
composed of elites who influence policy through the professionalisation of issues
(portraying policy as technical issues requiring expertise).
2. Supra-national institutions – obvious examples are the EU, OECD, World Bank
and UN. However, we can also identify the role of e.g. the UK government in
promoting transfer between regions. If we are being very specific, there was a
clear role for the old Scottish Office in helping Glasgow Health Board learn from
Lothian Health Board’s drug problem service.
Why Transfer - Is it Voluntary?
Perhaps the broader question may be: why change policy? Why innovate? Our
discussion in other lectures suggest that this begs a whole series of other questions, such
as: who changes policy? Who decides? Who influences? How much change would we
expect? Up until now, perhaps the “who decides” question assumed that we were
discussing a range of actors within one region. However, discussion of coercive policy
transfer may suggest that sources of influence may come from outside political systems.
Dolowitz and Marsh (1996: 346-8) outline 3 types of explanation (best seen as points on
a continuum rather than discrete categories):
1. Voluntary transfer – this ties in best with previous discussions. The short answer
is that decision-makers voluntarily engage in transfer following some level of
dissatisfaction with existing domestic policy. A sense of policy “failure”
(following a series of incremental steps?) produces an impetus for new policy
solutions. Bennett (1991: 200) suggests that there is a “natural tendency to look
abroad” to see how others have reacted to similar policy problems and to seek
ideas when innovation is required.
Maybe this account pre-supposes a level of rationality? If we focus on agenda-
setting we can say that a shift of attention was caused by some shock to the
existing system – change of government, successful reframing following new
evidence – and existing policy is now characterised as a failure. The shift in
power causes decision-makers to look outside of the failed organisation for new
ideas. Alternatively, as Dolowitz and Marsh (1996: 347) suggest, a new
government will look to international experience to legitimise its new aims. This
should arise when we discuss the global adoption of new public management.
2. Direct Coercive Transfer – this relates to influence (force?) by an organisation,
country or supra-national body to make another region adopt a certain policy. For
Dolowitz and Marsh (1996: 348; 2000: 10-11) direct influence by a country is
rare, although the role of the IMF or World Bank in setting policy conditions on
developing countries in exchange for loans is in turn influenced by individual
countries (most notably the US). Multi-national companies may effectively force
countries needing foreign direct investment to adopt liberal regulatory policies on
employment and environmental control (1996: 348).
Coercion is not restricted to developing countries. We can also point to the EU as
a “policy-pusher” through directives, while Bennett 91991: 228) highlights the
example of Britain lagging behind other countries on data protection and
threatened being “excluded from the ‘club’”. The power of business is also
apparent in the US – perhaps it is just not as visible when we look at transfer.
The power lectures will no doubt discuss Crenson’s study which suggests that
towns dependent on large single industrial businesses were less likely to press for
costly environmental policies.
3. Indirect Coercive Transfer – Dolowitz and Marsh (2000: 13) suggest that this is
voluntary but driven by a perceived necessity to change policy. In this case
region A may feel obliged to adopt the policy of region B, either because region B
is an important market for A’s exports, region B’s policy causes “externalities”,
regions A and B have an interest in working together and harmonising
arrangements (e.g. if they share borders and need similar environmental policies),
or if a region perceives the need to keep up with international policy
developments. Again, this has an interesting link with agenda-setting – in this
case (for example) domestic policy monopolies may be threatened by new policy
networks forming in other significant countries. Dolowitz and Marsh mention the
need for Canada to react to policy in the US. Since devolution, we also see the
need for Scotland and Wales to react to English developments. Wales is
particularly affected by policies on e.g. healthcare or university fees because there
is such a cross-border flow of people.iv
However, in case this seems too simple, they also note (2000: 11) that transfers may
contain voluntary and coercive elements, while levels of discretion (and hence
voluntarism) are inherent in most coercive forms of transfer. The perceived need to
transfer also varies over time according to the immediacy of the issue and its place on the
policy agenda. Perhaps a good example comes from one of Hoberg’s (2001: 127) factors
for convergence – “international economic integration”, in which the mobility of capital
and increase in global trade creates pressure or a feeling of pressure to minimise obstacles
to foreign direct investment or trade in a region. Hoberg (2001: 128) suggests that this
pressure has not translated to policy to the degree he would expect, suggesting that such
coercion is met by opposing pressures in individual political systems.
Another interesting point on coercion comes from Holzinger and Knill (2005: 781). We
should be careful about assuming a high level of coercion since e.g. there is a difference
between a government being resistant to a policy transfer and a government which sees
the policy as just not very high on its list of priorities. Further, the appearance of
coercion may help governments introduce policy “not favoured by its citizens”.
Obligated Transfer (transfer as a result of treaty obligations etc)
Lesson-Drawing Coercive Transfer
(perfect rationality) (direct imposition)
Lesson-Drawing Voluntarily but Conditionality
(bounded rationality) driven by perceived
necessity (such as the
desire for international
acceptance)
(Source: Dolowitz and Marsh, 2000: 13)
What is Transferred?
Bennet (1991) identifies similarities across a number of countries in terms of policy
goals, content, instruments, outcomes and/ or styles. Dolowitz and Marsh (1996: 349-50)
suggest that this is too limited, listing 7 things: “policy goals, structure and content;
policy instruments or administrative techniques; institutions; ideology; ideas; attitudes
and concepts; and negative lessons”. It is difficult to tell if the latter’s list is any better
since presumably some of the latter’s could be subsets within Bennett’s categories. There
is also some degree of overlap of categories – e.g. the different between policy goals and
instruments may be as difficult to disentangle as, say, policy formation and
implementation. The point is that – again – there is no straightforward answer here and
we should bear in mind that transfer does not just relate to “policy”. If a region has
already adopted the ideology or policy of another, it may be looking for more practical
advice on how to implement. There will also be interesting examples of region A
adopting the policy of region B even though there is an ideological gulf between them.
For example, in both England and Wales there is a commitment to offering patients the
chance to travel to another health authority if they have been waiting too long for
treatment. In England they call this “consumer choice” while in Wales they link it to
“citizen rights”. We can see a similar example in the adoption of new public
management in New Zealand by a left-wing government.
One issue to think about is: what does it mean if e.g. policy outcomes are the same in
certain regions but the processes that produced the outcomes are different? (see Bennett,
1991: 229). Does it suggest that there is a more important independent variable (policy
conditions, public opinion, the role of experts, resource constraints …)? We should see
this to some extent in the UK discussion of smoking policy convergence. All 4
eventually went for similar bans on smoking in public places, but the routes taken by
each administration were clearly different and the result far from inevitable. A related
point which is explored near the end relates to implementation: what does it mean if
regions adopt the same policy but very different means to ensure success?
Degrees of Transfer
For Rose (1993: 30) the extent of learning varies, from:
1. Complete duplication – Rose (1993: 30) suggests that this is only possible within
political systems (e.g. across US states) “because of the identity or close similarity
of institutions and laws”. Copying is more difficult across nations because of
differences in language and laws. Dolowitz and Marsh (1996: 351) suggest that a
way to prove copying is to find identical wording in the legislation. In this
context the UK example is interesting. There are various examples of Scottish
Acts following devolution which all but copy Westminster Acts (see Keating,
Stevenson, Cairney and Taylor, 2003). However, the wording has to be different
in many sections to take Scots law and administration into account. So is this
copying or adaptation? If it is the latter then this also applies pre-devolution!
2. Adaptation (taking different starting points into account) – this is a more likely
strategy for an importing country which has to adapt policy to local
circumstances. The link with incrementalism should be clear – policy change
cannot be wholesale or from a blank piece of paper.
3. Making a hybrid from the exporting and importing region – this seems very
similar to adaptation (although Dolowitz and Marsh suggest it is similar to
synthesis!). Rose gives the example of transfer from federal to unitary state, with
the policy adopted but the administrative arrangements maintained.
4. Synthesis - extracting some aspects of one or more exporting programmes.
Rose’s example is new democracies which borrow elements of electoral systems
and legislative models to make a new whole.
5. Using lessons from other regions as a broad inspiration from policy change. In
this case the substance may be less important than the chance it affords for policy
makers to see that things can be done successfully if done differently. The danger
of course is that if no detailed analysis takes place then the faults are not fully
studies and the conditions for success are not fully understood. If you study
Scottish Politics this may remind you of the example of using the “consensual
democracies” as an inspiration for Scotland’s political system.
6. Perhaps from other discussions we can add a sixth category at the other end of the
spectrum –using information gleaned elsewhere to merely repackage existing
policy. Dolowitz (2003: 103) suggests that this is apparent in some of the
motivation behind the child support agency transfer and the approach of the
Conservative party in opposition.
Four things should be borne in mind –
We can see the 5 types as on a continuum. Again, it does not do to get too
bogged down in the boundaries between them.
Rose (1993: 32) suggests that it is possible for all elements to be apparent within
one policy area. Rose gives the example of tax policy.
There is no assumption of the period of time in which transfer takes place – it
could be a single act or change over years (see Page, 2000: 5).
Presumably we have to bear in mind the extent to which policy would have
changed regardless of the transfer. For example, a change of government may
produce some sort of commitment to e.g. attempt to reduce unemployment by
reducing unemployment benefit entitlement. The government may then
completely import a policy from another region, but the transfer element may
only explain e.g. the administrative aspects or the levels of reduction.
From Where Are Lessons Drawn?
This discussion ties in with “Constraints on Transfer”. Rose (1993: 98-9) suggests that
the starting point for lesson-drawing is learning from the past in one’s own region. Then,
lessons from other regions are more likely if the exporting region shares similar:
policy conditions (particularly economic since there is a question of affordability)
geographical borders
ideology.
Again, the obvious example is between states in the US, although as Rose (1993: 99-102)
this is not straightforward since states have to some extent to choose which states are
“nearest”, while e.g. policy conditions for cities will differ from more rural areas.
Dolowitz and Marsh (1996: 350) give the example of ideological similarities between the
UK and US during the Thatcher/ Reagan era which perhaps made the UK predisposed
towards looking to the states for ideas.
Or, a country may be an exemplar with demonstrable expertise that catches the interest of
others. As Rose (1993: 107-8) discusses, Sweden has long been a beacon for the social
democratic state, Germany has a history of inflation control, and the US and now Japan
have become known for a degree of innovation. Of course, ideological commitment
influences the search for ideas since e.g. a country committed to rolling back the state
would not necessarily look to Sweden! Also remember the point from indirect coercive
transfer – smaller or less powerful regions may be effectively forced to look to their
larger neighbours since they know that significant divergence from that policy may have
consequences.
Perhaps we can take from Dolowitz and Marsh (2000: 7) that these factors (conditions,
borders) are less important over time as global economic forces and the rapid growth of
technology and communication, transport and the role of supra-national organisations
pushes regions towards more and more transfer “across diverse countries”.
Factors affecting the likelihood of transfer (and its success)
Rose (1993: 132) suggests that the attempt of and the success of policy transfer will
depend on a range of factors:
1. The policy should not be “unique” or depend on “inimitable” organisations
2. There should be few resource constraints to implement policy
3. The policy should be simple with a clear cause-and–effect
4. The smaller the departure form original policy the better the chances of success
5. Transfer is more likely when there is interdependence (think of Wales working
out whether to follow England)
6. Transfer is more likely when the values of the importer and the exporter coincide
7. Dolowitz and Marsh (1996: 350) also interpret Rose’s focus on the level of side-
effects of a policy, the amount of information gathered on the source policy and
the ease of prediction of outcomes.
8. Also remember the discussion on the importance of similarity in political
structures and administrative arrangements. For example, it may be difficult to
transfer policy from federal US to unitary UK (1996: 354). Dolowitz (2003: 106)
highlights this problem in welfare policy – any study of the US should recognise
that federal policy is pursued on the assumption of supportive welfare services at
state and local levels. There are also crucial differences in e.g. the potential role
of the private sector in delivering services in developing countries (Page, 2000:
10). However, there will also be issues within states – such as the applicability of
models to England and Wales even before devolution. In particular, policies
imported and aimed at fostering competition within public services were
inappropriate in Wales given the spread of services and geographical constraints
on service and people movement.
It should be clear from this list that what we are talking about is the applicability of
comprehensive rationality (particularly 2 and 7) and incrementalism (4), while the whole
list seems to follow closely the traditional conditions associated with top-down
implementationv. In other words, it is difficult to disentangle the discussion of transfer
constraints from general issues encountered in the existing literature. Perhaps only
conditions 5, 6 and 8 suggest something relatively new.
Policy Transfer and Policy Failure
The interesting thing about coercive policy transfer is that there is great potential to be
misled about the level or amount of policy that can be forced upon a country. A good
parallel is with the role of the UK Treasury in Scotland – we may argue that it is very
powerful in the sense that it can force other departments to save money, or it can
influence e.g. the actions on the Scottish Executive through financial incentives.
However, it is another thing altogether to suggest that the Treasury has the power to
successfully implement policy or to follow through policy to the “street” level. Surely it
is the same with coercive transfer – we may be seeing influence at the policy formation
phase without examining policy over a longer period or in more detail (scratching
beneath the surface and assessing the scope for detailed policy direction regardless of the
instruments of coercion available to e.g. supra-national organisations). One example is
the EU as a force for transfer. On top of Dolowitz and Marsh’s (2000: 15) point that
since nations joined the EU voluntarily it is difficult to talk of coercion (although see
below – D&M often use the term coercion to mean feeling obliged towards policy
change) we can suggest that e.g. EU directives are subject to scope in their interpretation,
while the penalties (e.g. fines) for non-compliance may not be particularly serious.
Dolowitz and Marsh (2000: 17) discuss policy failure in a different sense – i.e. the extent
to which a policy was adapted properly. This has three (not mutually exclusive) aspects
(which they demonstrate in discussion of the UK’s child support agency):
1. Uninformed transfer – when the borrowing country has incomplete information
about the crucial elements that made the policy a success in the lending country.
See also Dolowitz’s (2003: 105) suggestion that seeking information from those
with a vested interest in the original project may not provide the full picture.
2. Incomplete transfer – when those crucial elements of policy are not transferred
(e.g. the phase-in time and role of the courts)
3. Inappropriate transfer – when not enough attention is paid to adaptation, or the
differences in policy conditions and structures and also policy aims of the lending
country (such as targeting those in arrears, as opposed to those could afford to pay
the most).
Again, the potential for confusion with the implementation literature should be clear. In
particular, it is worth considering what incomplete transfer means. Bennett (1991: 230)
highlights convergence of policy on data protection “yet a divergence on the instruments
chosen to enforce them”. Is this factor included within “incomplete transfer” since the
means to implement are integral to the policy itself? If there is insufficient political will
devoted to implementation, is this a failing of the transfer process?
Remember the other lectures
This is getting repetitive now, but one of the biggest constraints on policy change (and
therefore transfer) is existing policy and the power relations which underpin the current
policy agreement. Remember that our discussions from the 1st 3 lectures all undermined
the notion of comprehensive rationality:
The identification of policy problems and their solutions cannot just be selected.
They are produced and as such are subject to definition or framing.
The policy process may be incremental in character with long periods of stability.
The assumption (within comprehensive rationality) of a powerful central force or
actor is undermined by the literature on multi-level governance.
However, agenda-setting analysis suggests that short periods of intense policy
change occur even in policy areas previously characterised as policy monopolies.
This involves a successful redefinition of policy problems at one or more level of
government.
The relevance to policy transfer should be clear: while it may be important to look at the
source (or exporters) of policy ideas (other countries, individuals, think tanks, etc), we
cannot assume that the importation of those ideas will be straightforward. Any policy
transfer will be subject to the same conditions – incrementalism, governance, agenda-
setting – as any other policy, regardless of the source. This is a key point by James and
Lodge (2003) who question the distinctiveness of policy transfer research when many of
the questions involved are covered in the existing literature on policy change. We can go
into this in more detail below but it is easy enough to see the potential problems for
policy transfer in terms of the rationality model:
Agenda setting 1- lessons from other countries are not just “there”. They have to
be reported and as such are subject to framing and evaluation. As Page (2000: 4)
puts it, “the perception of how the policy operated in the exporter jurisdiction
might be crucially shaped by the observer from outside who seeks to import it”.
A good example for lecture 6 is smoking, with particular debate in the UK over
the “success” of tobacco policies in Ireland and elsewhere. Other examples will
show us that the value of the original policy may be defined in a particular way by
the importer. E.g. Dolowitz points out that the original aim of the Child Support
Agency in the UK was to raise revenue (thus focussing on a small number of
absent parents) even though the policy in the US was geared more towards wide
coverage to a large extent regardless of ability to pay.
Agenda-setting 2 – this is apparent in the choice of which regions to study. E.g. a
government may focus its search on ideologically similar regions rather than a
more “rational” process based on geography or similarity in policy conditions.
Agenda setting 3 - the pressure to transfer policy will reflect the pressure to
change policy in general – and therefore on the extent to which the issue has risen
on the policy agenda. This relates to Dolowitz and Marsh’s indirect coercive
transfer.
Agenda-setting 4 – a big part of agenda setting is about maintaining policy
monopolies by framing issues and e.g. maintaining low levels of interest in
“technical” policies. Transfer or international experience could also be used or
repackaged to legitimise existing policy.
Incrementalism 1 - the focus on learning may be restricted to countries which
share basic characteristics with the importing country. The search for ideas from
countries with different starting points may be unrealistic given the scope for
radical change.
Incrementalism 2 – governments learn from their own mistakes and a large
preoccupation of decision-makers is solving problems caused by previous
policies. At the very least we should bear in mind that the search for outside
experience is not automatic. It requires a level of dissatisfaction with existing
policy that translates to action and influence of the relevant decision-makers.
Incrementalism 3 – see the example of path dependence in Page (2000: 6). Japan
begins to borrow Germany’s police system because it had already borrowed
related aspects in law and local government.
MLG 1- the adoption of policy from one level of government may be undermined
by or dependent on the cooperation from another. Bennett (1991: 223) alludes to
the difficulty in demonstrating the effect of emulation: “The difficulty lies in
comparing the relative significance of emulation with all the other influences
(domestic and transnational) that may affect the assumptions and perceptions of
the multiple actors who shape public policy”. Holzinger and Knill (2005: 776)
take a different tack, suggesting that policy transfer is only relevant to policy
outputs and not implementation, since the latter is subject to so much influence
from other actors and variables.
MLG and policy networks – Bennett (1991: 226) also highlights that policy
convergence may be fostered outwith the arena of central government. Rather,
harmonisation of policy takes place along professional or technical lines at a
lower level of government (as traditionally suggested in the policy networks/
communities literature). Perhaps a good example here is healthcare, in which
professional/ clinical links (e.g. at conferences or through journals) foster
emulation in day-to-day practices (of course there is a limit to this, through
national clinical guidelines).
This devolved nature of transfer makes measurements of divergence tricky, since
there is likely to be variation by policy area.
However, lessons from elsewhere may be a powerful tool for organisations or
governments promoting policy change. Again using the smoking example,
Scotland’s pursuit of a ban on smoking in public places was boosted by the
portrayed success in Ireland. The lessons from Scotland and prospect for change
in Wales and Northern Ireland changed the context for the debate in England. Of
course we can go too far with the effect of influence from Scotland et al on
England. See Indirect Coercive Transfer for a discussion on England’s
overarching influence.
Remember the issue of time – if transfer is a one-off then there is obviously
power vested in the individual or organisation charged with the transfer. If we are
discussing a longer period there is more scope for the traditional interests to
become involved. Remember this when reading e.g. Schattschneider.
Issues with the Literature
Remember the broader context
The suggestion from some of the literature is that we should not ignore what else we
know about the policy process when looking at transfer. For example, Dolowitz and
Marsh (1996: 355) warn against the assumption that policy transfer is open to anyone,
when in fact it is subject to the same power relations and limits to influence as any other
aspect of policy. Page (2000: 4) makes a similar point: “the difficulty is in unravelling
the precise contribution of one strand … [transfer] in the complex mixture of ideas,
issues, compromises and practices that go to make up ‘policy’”. Perhaps the added value
of the transfer literature is focussing on the level of coercive transfer to explain policy
decisions made in e.g. developing countries.
James and Lodge (2003) go further in arguing that it is difficult to distinguish policy
transfer/ learning from other types of policy making. They argue that the factors listed by
Rose resemble a version of comprehensive rationality (while the discussion of transfer
failure also resembles the focus of implementation studies and Dolowitz’s 2003
discussion does read like a plea for a comprehensive search for transfer information). As
a result of the lack of distinction, and because transfer seems to be defined so broadly,
James and Lodge (2003) argue that measuring the rise in transfer is problematic (see
Dolowitz, 2003: 107 “policy transfer … has been going on ever since one caveman saw
how another was more effective at hunting and adopted similar techniques”. Does this
suggest that all learning is transfer?). Evans and Davies (1999: 365) make similar points
on the problem of measurement and explanations for rises in transfer.
This ties in well with what I have been saying about the other lectures – this is all
connected and these theories/ models are only separated for ease of discussion. However,
it is useful to bear in mind the point by Page (2000: 4) that we should consider what it is
we want to explain. A focus on the existing literature moves us “closer towards a study
of intra-organisational power” when the object of study may be the source of policy ideas
and their applicability outside the source organisation or region. However, James and
Lodge (2003) seem to suggest that we can explain the adoption of ideas from outside
existing policy monopolies or regions without the policy transfer literature since plenty
of literature already exists (e.g. in comparative politics?). The significance is the role of
ideas and how they are produced and adopted, and/or the ways in which institutions
affect learning processes.
Rationality and Power
James and Lodge (2003: 184) suggest that the Dolowitz and Marsh (2000: 9) emphasis on
the “Why Transfer?” continuum conflates issues of perfect rationality (associated with
voluntary transfer) with issues of power - when really there are 2 explanations for policy
change which should be separated. Policy transfer can be entirely voluntary but not
rational, while the existence of bounded rationality does not necessarily suggest a level of
coercion to change policy. Part of the explanation for the conflation may be that
Dolowitz and Marsh (2000: 14) seem to use the word “coercion” to mean that policy
transfer is voluntary but that regions may feel obliged to keep up or transfer policy to
address a pressing need (see e.g. the discussion on p.15 on pressure to address
unemployment). The suggestion may be that the more there is pressure to do something,
the less time there is for a comprehensive process of solution searching. However, this
internal type of coercion is not suitably separated from a discussion of external pressures
in the process (the EU, World Bank, UN, etc).
Dolowitz and Marsh are also unclear about who or what is coerced and this reinforces our
need to apply theories like policy transfer to the existing literature. For example, let’s say
that we are discussing 2 advocacy coalitions and the dominance of one is replaced by the
dominance of another. For argument’s sake, let’s say that the dominance of an advocacy
coalition based on voluntary/ libertarian/ industry favoured measures to address the
effects of tobacco is challenged successfully by a public health coalition looking for
smoking cessation measures and a ban on smoking in public places. If policy changes as
a result – and on the basis of, say, international evidence and applying experience from
elsewhere – does this imply coercion? Does policy change by force? Surely this depends
on the position of the key decision-makers – as committed participants within coalitions
or as referees to a wider process. If the former, then maybe these decision makers were
“forced” to change their position on the basis of the process of agenda-setting and power
shift. If it is the latter, then maybe we have a “rational” decision-maker merely choosing
a policy alternative in the face of new evidence? Was the recent Westminster vote on
smoking based on a government coerced into policy change following changes in the rest
of the UK and elsewhere, or merely a voluntary policy change based on MP conviction in
a free vote? Again, the importance of coercion needs to be more fully specified. Who is
coerced? A key decision-maker? A government? A country/ region? Who is doing the
coercing? Is it pressure felt or pressure exerted?
There is also scope for confusion in Dolowitz and Marsh’s (2000: 16) claim that the
transfer of the American Earned Income Tax Credit as the Working Families Tax Credit
“falls as close to the purely rational end of the continuum as policy is ever likely to”.
This assertion is based on the time that the UK government had to study the issue and the
relative lack of pressure to change policy. However, as we saw in lecture 6.1, rationality
is not ensured just by a lack of pressure! There are insurmountable problems to
comprehensive rationality which do not relate to political pressure [in the context of
transfer, one huge obstacle seems to be the difficulty in determining the original goals of
the exporting country (Page, 2000: 10)]. From recent experience and reporting of the
WFTC we can see that the biggest problem has been the inability to anticipate the
problems of the policy itself – based on a lack of resources to implement and the ongoing
problem of incomplete and constantly out-of-date information when processing claims.
(note that Dolowitz himself disputes the idea that the policy has failed, and of course this
is subject to interpretation and media exaggeration – see http://www.hm-
treasury.gov.uk/newsroom_and_speeches/press/2006/press_13_06.cfm)
How is policy transfer demonstrated?
This may seem like an obvious point, but if policy transfer is important, it would be good
if we could show that it happens. James and Lodge (2003) criticise Dolowitz and
Marsh’s assertion that policy transfer is increasingly important since they base it either on
indirect evidence (such as the greater potential to transfer following the rise in
communication) or a handful of examples which are not as randomly chosen as D&M
would like us to believe. However, Page (2000: 4) suggests that we should not get
bogged down in precisely identifying the point of decision-making when transfer was
adopted (since this detracts from transfer and focuses on intra-organisational power).
Bennett (1991: 222) suggests that “observation of the utilization of foreign experience is
not proof of borrowing or copying”. There is also a need to look in detail at the
substance of laws and regulations for evidence of similarities.
A Final Note
If we focus too much on the qualifications and criticisms then of course we may start to
wonder: why bother? There is value to the transfer literature – it shows us where the
ideas for policy come from and perhaps if there is incomplete transfer it explains why
policies do not work as well in region B compared to region A. It is worth looking at the
case studies for more on thisvi and bearing in mind that the source of policy change may
be from an idea developed elsewhere. As Dolowitz (2003: 101) also suggests, policy
transfer is important to the student of politics because it is increasingly in the minds of
governments. However, the link to rationality should be clear and so it doesn’t hurt to
have some grounding in what are now classic debates.
References – for starters
Bennett, Colin J. (1991) “What is Policy Convergence and What Causes It?”, British
Journal of Political Science, 21, 2, 215-33
Dolowitz, David P. (2003) “A Policy-maker’s Guide to Policy Transfer”, The Political
Quarterly, 101-8
Dolowitz, David P. and Marsh, David (1996) “Who Learns What from Whom: a Review
of the Policy Transfer Literature”, Political Studies, XLIV, 343-57
Dolowitz, David P. and Marsh, David (2000) “Learning from Abroad: The Role of Policy
Transfer in Contemporary Policy-Making”, Governance, 13, 1, 5-24
James, Oliver and Lodge, Martin (2003) “The Limitations of ‘Policy Transfer’ and
‘Lesson Drawing’ for Public Policy Research”, Political Studies Review, 1, 179-93
Page, Edward C. (2000) “Future Governance and the Literature on Policy Trasnfer and
Lesson Drawing”, ESRC Future Governance Workshop 28 January 2000
Rose, Richard (1993) Lesson-Drawing in Public Policy (New Jersey: Chatham House)
320.6 Ros
Stone, Diane (1999) “Learning Lessons and Transferring Policy across Time, Space and
Disciplines”, Politics, 19, 1, 51-9
Further Reading
Bennett, Colin J. and Howlett, Michael (1992) “The lessons of learning: Reconciling
theories of policy learning and policy change”, Policy Sciences, 25, 275-94
Evans, Mark and Davies, Jonathan (1999) “Understanding Policy Transfer: A Muluti-
Level, Multi-Disciplinary Perspective”, Public Administration, 77, 2, 361-85
Hoberg, George (2001) “Globalization and Policy Convergence: Symposium Overview”,
Journal of Comparative Policy Analysis: Research and Practice, 3, 127-32
Holzinger, Katharina and Knill, Christoph (2005) “Causes and conditions of cross-
national policy convergence”, Journal of European Public Policy, 12, 5, 775-96
Sanderson, Ian (2002) “Evaluation, Policy Learning and Evidence-Based Policy
Making”, Public Administration, 80, 1, 1-22
Walker, J.L. (1969) “The Diffusion of Innovations among the American States”,
American Political Science Review, 63, 880-99
i
I will use regiona and country interchangeably. While we should bear in mind that transfer/ learning can
be from within countries, across regions and cities, between organisations and elsewhere, I will not point
this out each time!
ii
And we have encountered Rose’s argument on the limits to policy choice over inheritance (week 6.1).
iii
See Dolowitz and Marsh (1996: 345) on ombudsmen, O’Neill (2000: 63) on Professor Alan Enthoven’s
role in NHS reform, …
iv
The Welsh example is interesting since it shows we cannot discount the role of party politics even when
policy transfer seems inevitable. In particular, the Rees report (commissioned to help Welsh decision-
makers deal with the introduction of top-up fees in England) suggested that introducing fees was inevitable
– a significantly different policy would either mean too many Welsh students leaving or too many English
students moving to Wales and crowding out Welsh student places. The spanner in the works came when
the Labour administration was reduced to a minority (following the defection of Peter Law from the Party)
and the opposition took the opportunity to reject the Labour plan based on the Rees report.
v
(1) That there is an understanding of, and agreement on, clear and consistent objectives;
(2) That a valid/ adequate causal theory exists, in which the relationship between cause and effect is
direct (i.e. that the policy will work as intended when implemented);
(3) That subsequent tasks are fully specified and communicated (in correct sequence) to a team of
skilful and compliant officials;
(4) That the required time and resources (including political will) are available, and fully committed,
to the relevant programme;
(5) That dependency relationships are minimal and support from interest groups is maintained;
(6) That external, or socioeconomic, conditions do not significantly constrain, or undermine, the
process (this includes a broad notion of “unintended consequences”).
vi
Particularly since the theoretical discussions are fairly open-ended.