POLITICS BY HEURISTICS: POLICY NETWORKS WITH A FOCUS ON ACTOR RESOURCES, AS ILLUSTRATED BY THE CASE OF RENEWABLE ENERGY POLICY UNDER NEW LABOUR
| Author | DAVID TOKE |
| DOI | http://doi.org/10.1111/j.1467-9299.2010.01839.x |
| Published date | 01 September 2010 |
| Date | 01 September 2010 |
doi: 10.1111/j.1467-9299.2010.01839.x
POLITICS BY HEURISTICS: POLICY NETWORKS WITH
A FOCUS ON ACTOR RESOURCES, AS ILLUSTRATED
BY THE CASE OF RENEWABLE ENERGY POLICY UNDER
NEW LABOUR
DAVID TOKE
Policy network analysis is criticized for being a ‘heuristic’ device, yet ‘heuristic’ methods may be
essential to achieve detailed understandings of specific policy outcomes. Rationa l choice modelling
alone cannot perform a similar function. This paper develops a ‘heuristic’ policy network approach
that focuses on the analysis of actor resources. Changing contexts can alter the resource distributions
of actors within a policy community. This can lead to new policy outcomes. Policy networks can
therefore be rescued from criticisms made by, for example, Do wding, by re-visiting Rhodes’s earlier
emphasis on analysis of actor resources. This approach is illustrated in the case of UK renewable
energy policy under the UK government of New Labour. Changing contexts have strengthened
the resources of the main renewable energy interest groups to achieve higher targets and more
technology-specific means of financial incentives. The Renewable Energy Association has achieved
legislation favouring feed-in tariffs as is the practice elsewhere in Europe for small renewable
generators.
There appears to be a divide in political science between those who promote and those
who criticize the use of ‘heuristics’. This argument has especially affected the debate about
the usefulness and application of policy network analysis. Indeed, one recent paper, citing
Dowding 2001, talks about how ‘those who employ networks as a heuristic device cannot
offer an in-depth analysis of policy processes’ (Christopoulos 2008, p. 478). Dowding
argues that ‘description itself . . . [like policy network analysis] . . . cannot substitute
for the modelling process’ (Dowding 2001, p. 91). Following this argument, labelling
something ‘heuristic’ is a form of abuse. In this paper it is argued that far from heuristics
being an inferior form of approach, it may be the main or perhaps the only practicable
way of deep analysis of specific public policy case studies. However, such objectives
may be effectively pursued in the context of combining the original, simple, Marsh and
Rhodes (1992) policy networks formulation in tandem with the original Rhodes (1981,
1999) emphasis on how actors use resources.
The aim of this paper is to demonstrate, using the example of renewable energy
policy, how such an actor-resource-centred policy networks approach can constitute a
useful heuristic method. In doing so there will be a focus on how changing resource
distributions among actors can help explain policy outcomes and how these changing
resource distributions may be associated with changes in the policy context.
The lack of focus on actor resources argued by Dowding was a key element of his (1995)
critique of how policy network analysis failed to account for change. However, as argued
in this paper, this problem can be effectively dealt with by revisiting Rhodes’s original
emphasis on actor resources as a way of explaining policy outcomes in policy networks.
This ‘original emphasis’ is contained in Rhodes’s ‘power dependence’ framework which
was initially devised to analyse policy relationships between central and local government
(Rhodes 1981, 1999).
David Toke is in the Department of Politics and International Studies, University of Birmingham.
Public Administration Vol. 88, No. 3, 2010 (764–781)
©2010 Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden,
MA 02148, USA.
POLITICS BY HEURISTICS 765
It is now necessary to discuss a policy network framework, since this will define the
relationship between the actors and also be associated with the sort of policy outcome
decision that may be reached. This paper introduces and uses the policy network
framework developed by Marsh and Rhodes (1992). The advantages of this approach
include its relative simplicity, widespread applicability, and purchase in analysis of
specific policy-making cases. We also need to examine key changes in the context because
this may influence how the actors perceive their own self-interest. Crucially, we need
to analyse how the resources and policy preferences of the actors are altered by the
context since this will have a crucial bearing on the outcomes. Contextual changes can be
associated with changes in the resources available to the actors in the policy networks.
This can alter the actors’ ability to influence the policy outcomes.
First, the paper will engage in a discussion about the roles of heuristics and modelling.
Then an approach to using policy networks will be set out, followed by the development
of a means of analysing actor resources. This method will be applied to the case study of
British renewable energy policy. This discussion is important for political science for at
least two reasons. First, it develops an important debate about the roles of policy network
analysis and heuristics in political science. Second, this topic represents an important
empirical topic involving an intersection of energy and environmental issues.
THE ROLE OF HEURISTICS
One definition of heuristics is that it is ‘solving problems approximately that cannot
be solved exactly’ (Journal of Heuristics http://www.springer.com/math/applications/
journal/ 10732, accessed June 2008). Another definition says that a heuristic device is ‘Of or
relating to a usually speculative formulation serving as a guide in the investigation or solu-
tion of a problem: ‘‘The historian discovers the past by the judicious use of such a heuris-
tic device as the ‘‘ideal type’’’ Karl J. Weintraub (http://www.thefreedictionary.com/
heuristic, accessed June 2008).
These definitions may flow from the use of the term by Lakatos (1976) who argued
that mathematicians can use approximations as an initial way of thinking about whether
hypotheses can be proved: ‘the incessant improvement of guesses by speculation and
criticism’ (Lakatos 1976, p. 3). Political science is not mathematics, of course, and in
political science one type of heuristic may be a means of classifying concepts to analyse
the detail of specific cases of public policy decision making. It may be possible to propose
generalizations that may later be inspected for their relevance to other case studies.
By contrast, behaviouralists and rational choice analysts argue that formal models offer
the possibility of prediction and generalization of laws of political behaviour. The formal
methods or models are meant to be representations of real world situations. Such methods
can involve hypothesis testing through quantitative methods or, in the case of rational
choice analysis, empirical testing of predictions based on game theory modelling. These
methods can be important in many types of analysis, although the extent to which they
generate universal laws is philosophically controversially. However, as this paper argues,
the use of such formal models in understanding the political outcomes of public policy
battles is limited. This paper now analyses debate (mainly between Dowding 1991, 1995,
2001 and Marsh et al. 1992, 1998, 2000, 2001) about the relative utility of policy network
analysis and rational choice modelling.
Public Administration Vol. 88, No. 3, 2010 (764–781)
©2010 Blackwell Publishing Ltd.
To continue reading
Request your trial