A System Perspective for Policy Analysis and Understanding: The Policy Process Networks

A System Perspective for Policy Analysis and

Understanding: The Policy Process Networks
Yael Parag
Visiting Research Fellow in the Environmental Change Institute (ECI), Oxford
University Centre for The Environment (OUCE)

Abstract

This paper introduces the Policy Process Networks (PPN) as a new framework for policy understanding and
analysis. PPN combines the strengths of the Policy Cycle and the Policy Networks’ approaches with the
systems thinking ideas of dynamic process and interdependencies. I argue here that different actors’
networks participate in the different stages of the policy process and influence each other as well as the
stages’ outcomes. Therefore, in order to better understand how policies are shaped, why, and by whom, a
PPN methodology is introduced as a mean for a systematic and comprehensive policy analysis. This
methodology is utilized in a case study explaining the formation of Air Emissions policy in Israel.

Key Words:
Policy analysis, Policy networks, Policy process, Policy cycle, Environmental Policy, System thinking.

Introduction

Public policies are one of the main means through which order is set in societies and large systems, such
as states, are governed. Public policies also play a key role in introducing changes to societies and in
altering individual and collective behaviour. Therefore, analysing the process through which public
policies are shaped and implemented and detecting it’s strengths and weaknesses are the first steps for
understanding how we may design policies that improve order and governance and bring to an effective
change. Seeing policymaking processes as soft systems (Mulgan 2001), this paper suggests a new
framework for policy analysis and understanding, which utilizes the systems thinking ideas of dynamic
processes and multiple interdependencies.
Policymaking is a complex ongoing process that stretches over long periods of time and involves many
interests and participants, which may vary along the course of time. Policies are context influenced and
are embedded in national, economical, political, cultural, and social structures and contexts. As a result,
policies, like soft systems (Checkland 1981), are extremely actors-context-sector-site-issue dependent
and specific.
In order to better understand the complex process of policymaking, and to improve the process of policy
making itself, much effort is dedicated to policy analysis. The policy literature suggests a plethora of
perspectives and frameworks for policy analysis. The Policy Cycle framework is one such framework. It
aims to disaggregate the complex phenomenon of policy formation into manageable steps (Bridgman
and Davis 2003). It suggests breaking down the process into its sequence stages and examining what
happens in each stage separately while assuming that one stage influences the following (Howlett and
Ramesh 1995).

1
Yael Parag

The Policy Networks (PN) perspective offers a different way to tackle some of the complexities
involved in policymaking processes. Focusing on the meso-level, it considers the effect of both
governmental and non-governmental actors on the policy process. PN perspective concentrates on the
cluster of interests in the process as well as on the relations between the actors who participate in the
policy process - the network - and seeks to explain policy outcomes by these characteristics (Marsh and
Rhodes 1992). Cooperation, patterns of information flow, joined strategies, as well as other
characteristics of the actors’ inter-relations in the network are the main mean through which PN aim to
explain policy outcomes.
Yet, both perspectives ignore some important aspects of policymaking and are limited in their ability to
explain and predict policy outcomes. Combining systems ideas of interdependency and dynamics
(Richmond 1993) with both the Policy Cycle perspective and PN perspective, this paper suggests a new
framework for understanding the policy process. I argue here that each individual policy process
actually involves - and is therefore governed by - several different networks: the Policy Process
Networks (PPN). Each stage of any policy process is governed by a specific network – the Stage
Network. This network structure and characteristics are shaped by the institutions and the procedures
that govern the stage, and by the interactions between actors who have interest in the specific stage and
who have access to relevant decision making forums. Furthermore, each stage network operates in the
context of, and in relation with other stages’ networks. The outcomes of each policy stage can be
explained by its network characteristics and by opportunities and constrains imposed by other stages’
networks. Following the logic of the PPN perspective, this paper suggests a new methodology to carry
out case studies which enables analysing the policy formation as an ongoing, dynamic, interdependent
and in context process.
The rest of the paper continues as follow: The first section briefly discusses the Policy Cycle and PN
literatures and its main drawbacks. The second section presents the concept of PPN as a new system
thinking driven perspective for understanding policy formation and as a new methodology for policy
analysis. The third section utilizes the PPN methodology for analysing and explaining the formation of
Air Emissions policy in Israel. The concluding section suggests the potential of PPN perspective for
future policy research.

Policy Cycle

The Policy Cycle framework suggests that the policy process develops along sequential logical stages of
problem solving, in which policy decisions are made by decision makers, with one stage informing the
next (Bridgman and Davis 2003). “Decision-making” claim Howlett and Ramesh (2003:162) “is not a self-
contained stage, nor is it synonymous with the entire public policymaking process. Rather it is a specific stage
rooted firmly in the previous stages of the policy cycle”.
Albeit variations, the policy cycle usually includes the following stages: Agenda setting, problem
definition and analysis, policy tools selection, implementation, enforcement and evaluation (Howlett
and Ramesh 2003). It is widely agreed that policy cycle, as a framework, is an ideal type from which
every reality curves away (Bridgman and Davis 2000, Howlett and Ramesh 2003). In actuality, different
stages are sometimes shaped simultaneously and there may be a succession of feedback loops (Hill
2005).
The rational for observing each stage separately – which is similar to Bertalanffy (1968) hard systems
logic - is that each stage differs from the others in the sort of activities it involves, the expertise it
requires, and the procedures that govern it. Another important insight contributed by the stages model is
the understanding that different stages provide different sets of outcome, which affect other stages, even
if they occur simultaneously. Differentiating between the stages activities, procedures, outcomes, and
other characteristics reveals that the interests in shaping each stage outcomes vary between actors and
affect the actors’ participation.

2
A System Perspective for Policy Analysis and Understanding: The Policy Process Networks

A critique of the policy cycle as a framework for policy understanding highlights three main issues: (1)
its lack of theoretical ability to predict policy outcomes (Jenkins-Smith and Sabatier 1993), (2) its
notion that public policies are dominated and led by administrators rather than by other actors (Jenkins-
Smith and Sabatier 1993, Colebatch 2005), (3) its focus on the bureaucratic process while disregarding
content and context aspects (Everett 2003) and intergovernmental relations (Jenkins-Smith and Sabatier
1993). As a model for policy analysis, it also adheres too much to the normative process and for
imposing schematic stages on what actually happens (Hill 2005). Despite criticism, this concept is
useful for disaggregating the web of policy transactions and for examining the process through which
policies are made (DeLeon 1999, Pielke Jr. 2004)

Policy Networks

The Policy Networks (PN) approach emerged as a meso-level framework for policy theory and analysis.
It is based on the understanding that policies are not shaped solely by governmental agencies but are
rather outcomes of some sort of interactions and relations between governmental agencies, private
sector actors and civil society actors. PN usually include “all actors involved in the formulation and
implementation of a policy in a policy sector. They are characterized by predominantly informal
interaction between public and private actors with distinctive, but interdependent interests, who strive
to solve problems of collective action on a central, non-hierarchical level”(Borzel 1998: 260).
The PN perspective considers not only actors but also factors like economic characteristics and market
changes. Changes that occur in one or more of the arena dimensions - such as institutional, ideological,
economical or technological - can alter the network’s structure, and, in turn, influence the policy
outcomes (Marsh and Rhodes 1992). Therefore, this perspective is preferred as an analytical perspective
when dealing with complex policy issues that involve many actors and interests, such as environmental
issues, owing to its ability to capture multiple coexisting interests that are presented by different actors
in the policy arena (For example: Van Bueren, Klijn and Koppenjan 2003).
A wide agreement exists among many policy scholars that networks affect the policy process and its
outcomes (Marsh and Rhodes 1992, Marsh 1998). Yet, this perspective is criticized for being largely
descriptive rather than predictive (Mills and Saward 1994, Dowding 1995, McPherson and Raab 1988).
Other PN critiques emphasise the missing linkage between network models and models of the policy
process (Sabatier 1991, Peters 1998), and on the lack of attention to the dynamics that motivate actors
within the network and acts as a catalyst to the process (Peters 1998).
Indeed, the networks literature fails to capture the dynamic aspect of policymaking processes, the
manner through which one stage leads to the following, and the interdependencies between the stages.
PN literature tends to explain policy outcomes by focusing on the characteristic of one dominant
network and fails to differentiate between policy stages. Consequently, it fails to capture how the
network’s components and structure change – sometimes completely – along the policy process. It tends
to neglect the fact that different actors have different access to policymaking forums in the various
stages. It also overlooks the modifications that elements – such as actors’ interests, power positions,
resources, know-how, and legitimacy – go through from stage to stage. Using systems thinking
terminology, I argue that the PN literature (1) lacks a dynamic dimension of the policy process and, as a
result (2) fails to capture the interdependencies between networks. Ignoring these two aspects the PN
ability to explain both the policy process and its outcomes is rather limited.

The Dynamic Dimension

PN literature has failed to capture the dynamic dimensions of elements in the policy process. It also has
difficulties in capturing the modifications of interdependencies and actors relations along the process.
Studies using the PN framework usually emphasize only one stage of the process. Many of them focus
on the impact of PN on the tools selection stage (e.g. Saward 1992, Daugbjerg 1998, Gonzalez 1998,

3
Yael Parag

Kitchen 2000, Howlett 2002), others on the implementation stage (e.g. Brinkerhoff 1996), and few on
the problem analysis stage (e.g. Starik and Heuer 2002). But none of these studies look at the
policymaking process through all its stages. As a result, some very important process’ contextual
aspects are neglected.
As noted by the Policy Cycle advocates, different actors participate in the different stages. Therefore it
would be only logical to assume that the participating actors weave different alliances, cooperation
practices and networks in each stage, according to their power positions, resources availability and
interests. As a result, along the same policy process we should distinguish between several networks
operating in different stages. Nonetheless, different networks do not necessarily suggest different
composition of actors, but rather hint to variations in, for example, inter-actor relations, power
delegation, resource allocation, and information flow. Hence, I argue that policy analysis, as well as
policy theory, should pay attention to the dynamic dimension of the policy networks along the policy
process.

Networks Interdependencies
Policies are shaped in a variety of coexisting multiple institutional-cultural-economical contexts that
have a great influence on networks characteristics and thus on the networks effects on policy. Yet,
scholars claim that one contextual aspect that is often neglected, and thus not well theoretically
evaluated, is the multi co-existing networks that affect each other (Marsh 1998, Marsh and Smith 2000,
Van Bueren et al. 2003), or in other words: networks interdependency. In a complex polity, argue Marsh
and Smith (2000:8), the relationship between networks is clearly crucial. Hence, exploring these
relations can explain some policy outcomes. But Marsh also points at the different networks that operate
within the same policy process, influencing one another (Marsh 1998). He claims that it is possible that
actors who are involved in the policy tools selection networks are not necessary involved in the
implementation networks and that this can explain some implementation gaps (Marsh 1998:192).
Following Marsh’s notion regarding network-network relations within the same policy process, I argue
that in order to explain and understand policies, as well as to design policy effectively, all the networks
involved within the same policy formulation process should be considered. Due to the evolving manner
of policy formation process, in which the process can move backward and forward, it is important to
examine the network-network interdependencies.

Policy Process Networks

Consequently, I suggest the Policy Process Networks (PPN) perspective for policy analysis to tackle the
abovementioned weaknesses in the PN framework. The PPN relies on the Policy Cycle notions that (1)
different actors may participate in different policy stages, (2) different sort of activities and expertise are
required in each stage, and that (3) each stage’s outcomes affect other stages. From the PN literature it
adopts the notions that (1) policies are shaped by networks of state and non-state actors, (2) networks
should be examined in their contexts, and that (3) policy outcomes can be explained by network
characteristics. From the Systems thinking it adopts the ideas of (1) dynamic processes, and (2)
interdependency.
Networks dramatically vary along the policy formation process. It is not only that some actors
participate in some stages while abstaining from other, but also that actors who have a great influence
on a specific stage network, have a minor influence on another stage network (see for example: Perkin
and Court 2005). Thus, the PPN perspective suggests understanding policy formation by considering all
the different networks that participate along the process. While it enables us to look at the process
through separate stages it also provides a better understanding of the policy as a whole.

4
A System Perspective for Policy Analysis and Understanding: The Policy Process Networks

PPN Methodology for Policy Analysis

The application of a PPN framework also demands a new methodology, which enables analysing the
policy formation as an ongoing, dynamic, interdependent and in context process. A methodology that
enables us in a systematic manner to observe the various networks operating in the same policy process,
to understand the relations between these networks and to connect between networks and outcomes.
I suggest the following methodology. Its steps are the following: First, the policy process is broken
down to its sequence stage (i.e. agenda setting, problem definition, problem analysis, policy tools
selection, implementation, enforcement and evaluation). Second for each policy stage (1) the essential
resources are identified (2) the set of outcomes is detected (3) the network is identified and examined
(4) the outcomes are explained by the unique stage’s network characteristics, and (5) the network
characteristics are viewed and explained in the context of the other stages networks.
PPN analysis results in a detailed, structured, and in-context analysis, which emphasises the dynamic
aspects of networks along the process. Multiple memberships in networks, variations in inter-actor
relations and changes in flow of information patterns may shed light on some unnoticed aspects of the
policy process. Furthermore, PPN analysis enables to detect more accurately weaknesses and strengths
in the process, which in turn may contribute to designing policies more effectively. The following
demonstrates the PPN usefulness as a policy analysis tool.

A Case Study: PPN and the Israeli Industrial Emission Policy

Like most developed countries Israel has legislation that defines the maximum level of pollutants, which
are allowed in the outside air (Ambient Air Quality Standards). Usually, a complementary regulation for
air quality is provided by the Emission Standards, which establish the maximum amounts of pollution
that a given source is permitted to emit.
The Israeli Ambient Air Standards are aligned with the World Health Organization recommendations.
But these air standards are not accompanied with legislation that controls emissions to the air. Instead, a
voluntary agreement (the Treaty) exists, which controls only emissions from industrial sources. The
Treaty was signed in 1998 between the Israeli Ministry of the Environment (ME) and the Manufacturers
Association of Israel (MAI), which is the industrialists’ umbrella organization.1 The Treaty adopts
German and European Union industrial emission standards. Since it was issued, most of the Israeli
major polluting businesses have signed the Treaty. And yet, despite the Treaty, some of the ambient
standards are exceeded almost daily. One reason for this is that businesses that signed the Treaty often
violate it (Parag 2005).
Simple PN analysis would presumably focus on the ME and MAI strong cooperation and would explain
air quality in Israel by the governance of such an alliance. Yet, focusing on the ME and MAI network in
the national context per-se provides an incomplete picture and fails to explain why the Israeli air
emissions policy was formulated as it was and why the Treaty does not seem to be the ‘end point’ of the
process. A PPN analysis, which considers the dynamic dimension of the policy process and network-
network context, reveals, for example, that the relations between the ME and the MAI vary in the
different stages. The following sections use the PPN methodology for analysing this case2. Findings are
summarized in a table in Annex 1.

1
The Israeli Industrial Emission Treaty:
http://www.sviva.gov.il/Enviroment/Static/Binaries/Articals/0052_0.pdf.
2
The following case study is based on a variety of data sources collected in Israel between the years 2000 and 2004: semi-
structured interviews, publications, environmental conferences, parliamentary committees’ protocols, and official archives.

5
Yael Parag

Agenda setting
In order to promote the problem of air pollution and its related health effects pro-environmental actors
cooperate among themselves. The ME and some Environmental Non Governmental Organizations
(ENGOs) are the most active actors in the agenda setting stage. An ongoing information flow exists
between them. They support one another in activities that promote social and governmental awareness
to the air pollution problem. One example for this cooperation is the joint seminars they conduct to
various audiences about bad air quality and its effect on human health. These actors also support the
formation and maintenance of local environmental grassroots organizations: the ME provides some
money and the ENGOs information and ‘know how’. Severe pollution events that catch the media
attention are picked up by the ME and ENGOs and are used as a lever to increase public and local
authorities support for reducing air pollution.
As another example for such cooperation, The Israel Union for Environmental Defence (IUED), one of
the leading Israeli ENGOs, with support of the ME published a comprehensive report on the health
damages and their related economic costs caused by air pollution in Israel. This report3 suggests that
more than a thousand people die in Israel every year due to air pollution. Using public relations and
lobbyists, the ME and the IUED promoted this report in the media and in the government. The figures
presented gained high attention in the public discourse resulting in a number of the Parliament’s Interior
and Environmental Committee meetings that were dedicated to this topic and that called the government
to act to reduces air pollution.
By sharing information and cooperating with each other the ME and the ENGOs succeeded in raising
the profile of the air pollution problem in the national agenda. The need for further air pollution
reduction and the absence of mandatory regulation that control air emissions was discussed in the
media, in local authorities and in the parliament.

Problem Definition and Analysis

In the policy definition and analysis stage the network is restricted to those actors who present studies
regarding cost-benefit, risk, and health analysis. The participants in the network are several ministries
and the ENGOs. Although all the actors agree that pollution damages health they present different ways
to measure the related costs.
The IUED and the ME presented together an analysis related to the health damages and its associated
economic costs. They also presented an analysis, which focuses on the monetary benefits that
industrialists gain from not implementing technical solutions for pollution reduction (Tal 2002).
Industry, on the other hand, provided its own calculations of economic costs, emphasizing the huge
investments required which would prevent it from competing in the global market and eventually would
result in loss of jobs and an increase in unemployment. At the same time, together with the ME, the
industrialists presented the fact that most of the air pollution comes from power plants and
transportation. Therefore they both claim that any air pollution reduction policy should consider all
sources and not only the industrial ones.
The Ministry of Infrastructures is another actor in this network – a dominant one. This Ministry decides
on the grade quality of the oil Israel purchases. The better the oil grade, the less pollution it produces. It
is also the Ministry in charge of Israel’s Electricity Company. It is no surprise, then, that in its analysis
this Ministry underlines only the costs of emission reduction and its effects on the economy. This sort of
analysis was supported by the Ministry of Transportation that emphasised the anticipated increase of
fuels rates, which in turn will result in an increase of cost of goods. Both of these ministries claimed in
the decision making forums that the related extra costs would eventually be paid by the public.

3
A Comparative Assessment of Air Pollution Public Health Risks in Two Israeli Metropolitan Areas.
http://iued.org.il/text_item.aspx?tid=86&menu=7

6
A System Perspective for Policy Analysis and Understanding: The Policy Process Networks

Unlike the agenda setting network, in this stage network the ME is a relatively weak actor, compared to
the other Ministries. The information flow and data sharing between the ME and ENGOs, and even
between the ME and the MAI, is not enough. With no strong analytical support from the Ministry of
Health, decision makers are provided with health damages’ analysis originating from the ENGOs and
with economic analysis presented by strong Ministries who predict an increase in consumer goods cost.

Tools Selection
The first policy tool the ME turned to was overall emission legislation. However, the power distribution
within the tools network made legislation impossible. The dominance of economic considerations, the
power of the Ministry of Infrastructure and the Ministry of Transportation, which were backed by the
Ministry of Finance, versus the low priority of public health issues and the ME’s political weakness
were all demonstrated when the ME tried, and failed, to pass the regulation in the Israeli Parliament.
Accordingly, the ME tried to overcome its relative weak position in the network by avoiding
confrontation with strong actors and by turning to an alternative strategy – a strategy that places it in a
relatively strong power position. Fragmenting the problem to its different sources (industry,
transportation and power plants) was the first decision the ME took. Cooperating initially with the group
that has an interest in controlling emissions was the second.
The industry has interests in an arrangement that sets emissions standards because otherwise the ME
may use its only available tool to reduce polluting emissions – The Business Regulation Act (1968).
According to this Act the Ministry’s inspectors can refuse to approve business’ license if it does not
reduce its emissions. Yet, this tool was not designed to control emissions in the first place and therefore
has many limitations in doing so (Keret 2004). The MAI claims that the standards set in the licensing
procedure are unsystematic and may change unpredictably from one year to the next. This
inconsistency, in turn, increases uncertainty regarding required investments and impedes the
industrialists’ ability to plan for the future.
Sharing the same goal – establishing emission standards – although with different motivations, the MAI
together with the ME looked for an alternative policy tool for controlling emissions. A tool that will not
require any powerful Ministry agreement and that could be applied within a reasonable timescale.
Together they decided to adopt an arrangement existing in the Netherlands –a Voluntary Treaty. To this
Treaty framework they incorporated German and EU industrial emission standards. Upon voluntary
joining the Treaty these standards are incorporated into the business license issued by the ME. This
results in the business legal obligation to conform to the Treaty requirements. Among the obligations
are the requirements to implement a self-monitoring procedure and to report the results to the ME. A
Joint Implementation Committee was established to resolve disagreement and to update standards. Both
sides are members in the committee but the Ministry holds the majority.
The industrialists’ willingness to cooperate can be explained, as mentioned above, by the need for
certainty. But cooperation with the regulator also provides them with exclusive access to the decision-
making forums and with participation in constructing the policy tool so that it better suits their own
interests. From the Ministry’s point of view, signing this voluntary Treaty was a step forward toward the
goal of reducing air pollution.
Few large ENGOs, which in previous stages cooperated with the ME, confront it this stage. They claim
that the Treaty violates the rule of law, since as a voluntary agreement it is based upon industry’s
preferences, and such polluters lack an incentive to create strict standards. In addition, they claim, the
public – which is affected by this arrangement – has no access and no say in the process (Karo Yefet
and Papay 2001).

7
Yael Parag

Accordingly, the IUED appealed to the Supreme Court against the ME claiming that all emissions to the
air should be regulated by legislation.4 The IUED, has drafted a Clean Air bill and is lobbying for its
acceptance in the Israeli Parliament.5 In many senses, the IUED is attempting what the ME failed to do.
The ME cooperation with the MAI can be considered as an alternative exclusive network that shape the
tool and in which the Ministry is dominant. Yet, as part of the agreed cooperation, some important
actors are excluded from this alternative network. For example ENGOs, public representatives and
independent experts are not allowed to participate in the Implementation Committee, which revises the
Treaty on an ongoing basis. As a result, the representation of civil society interests is insufficient. At the
same time, it should be noted that the Treaty as a strategy enables the ME to tackle the industrial
emissions problem with limited resources by cooperating with industry instead of confronting it.

Implementation and Enforcement

The policy formation process does not end when policy tools have been chosen. The success of the
chosen policy is an outcome of proper implementation and enforcement.
As of January 2005, out of about 1700 enterprises members in the MAI only 150 signed the Treaty.
Most of these 150 enterprises used to be major polluters and have subsequently dramatically reduced
their emissions. Yet, despite the significant reduction, random checks that the ME carried out
consistently show that many of the enterprises examined were exceeding the standards set by the
Treaty.6 Moreover, the self-monitoring emission results provided by the enterprises themselves differ
dramatically from the random check done by the Ministry.
The abovementioned Implementation Committee governs and support implementation. Yet, unlike the
previous networks, where the industry was represented by the MAI in the negotiations with the ME, in
the implementation stage the network is much wider and more complex. It involves various businesses,
which are not necessary willing to cooperate with the ME like the MAI did. These businesses, with only
advisory support of the MAI have to implement the new standards into their manufacturing processes.
They are required to search for new technologies that will enable them to reduce and monitor emissions.
However, they have to do it with no financial support from the Government.
Whatever the reasons are for the poor implementation, it is obvious that no deterring enforcement
mechanism exists. Although the ME publishes its random checks results, which show that many
businesses emit more than allowed, since the introduction of the Treaty and up till 2004, the ME issued
only two indictments against offending enterprises (Keret 2004). Instead, the ME prefers to negotiate
with the offenders and to settle things out of court (Parag 2005). Eventually, polluting pays off because
no real economic incentives are used to prevent poor performances. But by avoiding the court and by
negotiation with each business, the ME also decreases transparency in the enforcement.
Israeli ENGOs usually hold a key role in implementation and enforcement of environmental policies
(Weinthal and Parag 2003). But the Treaty limits their ability to participate in both the implementation
and the enforcement networks. Since the Treaty is voluntary ENGOs do not have legal standing to
appeal to the courts against offenders (Keret 2004). Since no legal action is taken against offenders by
ENGOs or the ME, the Judiciary’s role in the industrial emission policy process, which could have been
significant, is marginal.

Evaluation
Both actors who manage the Treaty evaluate it as successful and want to maintain it. When the Treaty
was introduced in 1998, the ME considered it only as an intermediate arrangement on the way to
legislation. However, as time passed the Ministry changed its mind and has not done any effort to

4
No verdict yet.
5
IUED website http://www.yarok.org.il/ (IUED presenting the bill for Israel’s Clean Air Act 2002, 27.05.02).
6
In 2001 41% of the enterprises examined exceeded the standards, in 2002 60%, and in 2003 58%.

8
A System Perspective for Policy Analysis and Understanding: The Policy Process Networks

promote an overall legislation in the Parliament. The ME claims that thanks to the Treaty, industrial
emissions to the air were significantly reduced. In addition, it argues that due to the disagreement of
other powerful Ministries - in other words, the power distribution in the tools selection network – any
attempt to promote a bill for overall emission control would take few years and in the end of the process
the ME would be forced to agree on a relatively low standards. Instead of legislation, which means
confrontation in a network in which it has a weak position, the ME is now negotiating directly with the
Electricity Company – another major polluter - and wishes to introduce one more voluntary agreement
based on cooperation for emissions reduction.
The MAI considers the Treaty as a successful arrangement because by being a powerful actor in the
alternative tools selection network it enables businesses to slowly adjust to emission reductions.
Furthermore, taking into consideration the enforcement network which does not present any deterring
punishment to offenders, MAI members are not threatened by heavy penalties. Hence, MAI and the ME
both protect the Treaty.
ENGOs, on the other hand, evaluate the Treaty’s success by comparing it to the hypothetical situation
where all emissions are controlled by legislation. Using the mass media and Parliamentary lobbyists
they are demanding the ME to confront other Ministries in the original tools network and not to give up.
Using public relations and media coverage, their policy evaluation in many senses is tightly linked to
the agenda setting network.

Summary and Conclusions

This paper followed the logic of soft system thinking that calls for investigation of complex phenomena,
such as policy formation processes, in their contexts and while considering the relations between
different elements composing and effecting the phenomena. It was argued and demonstrated here that
the same policy process is actually governed by different PPN, which operate one in the context of the
other, influencing one another. Breaking down the process to its stages while considering the context
reveals few previously unnoticed important insights and explanations regarding who shape the emission
policy, how and why. By emphasizing the dynamic aspect of the process and the networks
interdependencies, PPN enables better detection of strengths and weaknesses of the process, which in
turn may lead to policy capacity improvement and to an effective change.
The analysis uncovers some constrains, which hinder problem solving and impede decision-making.
More specifically, it points at the weaknesses of implementation and enforcement networks. The Israeli
ME needs to rethink both networks - in terms of structure, participation rules, power distribution and
resource distribution - in order to improve these stages outcomes and to enhance the policy
effectiveness.
In general decision makers may find PPN perspective useful when thinking about new policy or policy
alternation. Detecting and taking into account the multiple actors and interests, the possible linkages and
interdependencies between the stages’ networks, and the constrains that one stage’s outcomes pose on
the following may help to avoid some future policy failures.
PN framework is often criticized as lacking a theoretical ability to predict outcomes and explain policy
transformation (Dowding 1995). Yet, adding system thinking ideas and studying policy making from
the PPN perspective point to a new aspect of networks to theorize upon: the relation between the
different networks governing the same process. Elements such as multiple memberships in different
PPN or the dynamics of confrontation versus cooperation along the process may hold the potential to
explain policy change and to contribute to PN theory.

9
Yael Parag
Annex 1: Policy Process Networks analysis in the case of the Israeli air emission policy

Policy Stage Networks Networks characteristics Essential ME position in the Stage outcomes
members resources network
Agenda Setting ME, ENGOs, Open network. Strong Information, media Leading and dominant The air pollution problem entered the headlines
Parliament cooperation between the ME attention, public actor. and was discussed in several Parliament
committee, local and ENGOs. awareness. committees more than 5 times.
authorities.
Problem Health, risk, and Restricted participation, Analytical expertises, Weak actor in a network The problem was framed and analyzed by most
Definition and economic experts, dominance of economic access to dominated by economic actors mainly in its economic aspect. Health
Analysis Ministries of consideration. Cooperation information, political considerations. aspects were promoted by ENGOs and not by the
Environment, between the ME and the power. Ministry of Health.
Infrastructure, MAI and in the same time
Finance, MAI. between the ME and
ENGOs.
Policy Tools The weakness of the In the alternative network the Technical expertise. Weak actor in the cabinet A voluntary Treaty is the selected tool, which
Selection ME in the ME is a strong and leading Access to the network. Strong actor in controls only emission from industry. This tool
Government brings it actor. It is a close and very decision-making the alternative network. does not refer to the health analysis but rather to
to form alternative restricted network. forum. the economic one.
network with the
MAI.
Implementation ME, MAI, business Network with many Financial resources, Weak actor in a multi About 150 businesses signed the Treat. They
that signed the participants from different technical expertise, actor’s network, in which have to implement it with no support from the
Treaty. business. Its members do not know how. many members not ME or the government.
necessarily fully adhere to fulfilling the agreement. Implementation can be measured in terms of how
the MAI and ME many businesses sign the Treaty and how many
cooperation. Information reduced emissions.
flow is not reliable.
Enforcement ME, Courts. The ME is the ‘gate keeper’ Financial resources Poor in enforcement The enforcement mechanism is weak. Despite
ENGOs have no of this network. It tends not (inspection), resources and thus the fact that many businesses exceed their
access. to involve courts but rather deterrence, access to relatively weak actor that permits, no deterring penalties mechanism exists.
to negotiate with the data, law expertise. promote cooperation over The ME prefers negotiating with offenders
polluters. confrontation with instead of punish them. Courts are hardly ever
polluters. involved.
Evaluation ME, MAI, ENGOs. Network that includes the Data, information, A dominant actor, who The Treaty is evaluated as successful tool to
ME and the MAI, but is alternative solutions. supports the current tool, reduced emission by both MAI and the ME. The
challenged by ENGOs. and does not want to policy solves only one aspect of the problem.
Closely related to the agenda change it.
setting stage.

10
A System Perspective for Policy Analysis and Understanding: The Policy Process Networks

References

Bertalanffy, L. (1968) General System Theory: Foundations, Development, Applications. New York:
George Braziller
Borzel, T. (1998) ‘Organizing Babylon - On the Different Conceptions of Policy Networks’ Public
Administration, 76(2):253-273.
Bridgman, P. and G. Davis (2000) Australian Policy Handbook, 2nd edition, Allen and Unwin, Sydney.
Bridgman P. and G. Davis (2003) ‘What Use is a Policy Cycle? Plenty, if the Aim is Clear’ Australian
Journal of Public Administration, 62(3):98–102.
Brinkerhoff, D.W. (2006) ‘Coordination Issues in Policy Implementation Networks: An illustration
from Madagascar’s Environmental Action Plan’ World Development, 24(9).
Checkland, P. (1981) Systems Thinking, Systems Practice. John Wiley & Sons, London
Colebatch, H. K (2005) ‘Policy analysis, Policy Practice and Political Science’ Australian Journal of
Public Administration, 64(3):14-23.
Daugbjerg, C. (1998) ‘Linking Policy Networks and Environmental Policies: Nitrate Policy Making in
Denmark and Sweden 1970-1995’ Public Administration, 76(2).
DeLeon, P. (1999). ‘The Stages Approach to the Policy Process: What Has It Done? Where Is It
Going?’ In: Sabatier P. (ed.) Theories of the Policy Process. University California, Davis.
Dowding, K. (1995) ‘Model or Metaphor? A Critical Review on the Policy Network Approach’
Political Studies, 43(1):136-158.
Everett, E. (2003) ‘The Policy Cycle: Democratic Process or Rational Paradigm Revisited?’ Australian
Journal of Public Administration, 62(2):65-70.
Gonzalez, G. A. (1998) ‘The Conservation Policy Network, 1890-1910: The Development and
Implementation of "Practical" Forestry’ Polity, 31(2): 269.
Hill, M. (2005) The Public Policy Process (4th edition), Pearson-Longman.
Howlett, M. and M. Ramesh (2003) Studying Public Policy: Policy Cycles and Policy Subsystems,
Oxford University Press.
Howlett, M. and M. Ramesh (1995) Studying Public Policy: Policy Cycles and Policy Subsystems,
Ontario, Oxford University Press.
Howlett, M. (2002) ‘Do Networks Matter? Linking Policy Network Structure to Policy Outcomes:
Evidence from Four Canadian Policy Sectors 1990-2000’ Canadian Journal of Political Science,
35(2): 235-267.
Jenkins-Smith and Sabatier (1993) ‘The Dynamics of Policy-Oriented Learning’. In: Sabatier, P.A and
H.C. Jenkins-Smith (eds.) Policy Change and Learning. An Advocacy Coalition Framework.
Oxford, Westview Press.
Karo-Yefet, A. and N. Papay (2001) ‘Mediterranean Pollution from the Kishon River’. In Papay, N.
(ed.) Israel’s Coast: Coastal Organization Forum Report on the Mediterranean Coastline
(Hebrew). SPNI. p 55-60.
Keret, D. (2004) Complementary Approaches to Environmental Enforcement – An Israeli Perspective:
ISO 14001 and Air Emission Standards Covenant. PhD Dissertation (Hebrew). Tel Aviv
University.
Kitchen, L. (2000) ‘Environmental Policy and the Differentiation of Rural Space: An Actor Network
Perspective’ Journal of Environmental Policy and Planning, 2:135-147.
Marsh, D. (ed.) (1998) Comparing Policy Networks. Open University Press.
Marsh, D. and R. A. W. Rhodes (Eds.) (1992) Policy Networks in British Government. Oxford,
Clarendon Press.
Marsh, D. and M. Smith (2000) ‘Understanding Policy Networks: Towards a Dialectical Approach’
Political Studies, 48: 4-21.

11
Yael Parag

Marsh, D. and M. Smith (2001) ‘Debate: There is More than One Way to Do a Political Science: on
Different Ways to Study Policy Networks’ Political Studies, 49:528-541.
McPherson, A. and C. Raab (1988) Governing Education. Edinburgh, Edinburgh University Press.
Mills, M. and M. Saward (1994) ‘All Very Well in Practice, But What About the Theory? A Critique of
the British Idea of Policy Networks’ Contemporary Political Studies, 1:79-92. Belfast: Political
Science Association.
Mulgan G. (2001) ‘Systems Thinking and the Practice of Government’ Systemist 23:23-29.
Parag, Y, (2005) Policy Process Networks: The Formation of Environmental Public Policy in Israel.
PhD Dissertation (Hebrew). Tel Aviv University.
Perkins, E. and J. Court (2005) ‘Networks and Policy Process in International development: A literature
review’ Working paper 252, Overseas Development Institute, London.
http://www.odi.org.uk/RAPID/Publications/Documents/WP252.pdf
Peters, G. (1998) ‘Policy Networks: Myth, Metaphor and Reality’. In: Marsh, D. (ed.) Comparing
Policy Networks. Open University Press.
Pielke Jr., R. (2004) ‘What Future for the Policy Sciences?’ Policy Sciences, 37:209-225.
Raab, C. D. (2001) ‘Understanding Policy Networks: A comment on Marsh and Smith’ Political
Studies, 49 551-556.
Richmond, B. (1993) ‘Systems Thinking: Critical Thinking Skills for the 1990s and Beyond’ System
Dynamic Review, 9(2): 113-133
Sabatier, P.A. (1991) ‘Toward a Better Understanding of the Policy Process’ Political Science and
Politics, 24(1): 44-156.
Saward, M.(1992) ‘The Civil Nuclear Network in Britain’. In Marsh and Rhodes (eds.), Policy
Networks in British Government. Oxford: Clarendon.
Starik, M. and M. Heuer (2002) ‘Strategic Inter-Organizational Environmentalism in the US: A Multi-
Sectoral Perspective of Alternating Eco-Policy Roles’ Business Strategy and the Environment,
11(4): 221-235.
Tal, A. (2002) ‘The Benefits from Not Complying with Environmental Law: The Role of Economic
Analysis in Decision of Penalties for Polluters’ In Idelman, A., A. Tal and N. Ben Aharon (eds.)
Studies on Environment and Policy (Hebrew). Jerusalem Institute for Israel Studies: The Centre
for Environmental Policy Research. Jerusalem.
Van Bueren E., Klijn E. and Koppenjan J. (2003) ‘Dealing with Wicked Problems in Networks:
Analyzing as Environmental Debate from a Network Perspective’ Journal of Public
Administration Research and Theory, 13(2):193-212.
Weinthal, E. and Y. Parag (2003) ‘Two Steps Forward, One Step Backward: Societal Capacities and
Israel’s Implementation of the Barcelona Convention and Mediterranean Action Plan’ Global
Environmental Politics, 3(1):51-71.

12