308 SAJEMS NS Vol 5 (2002) No 2
Economic Policy Making for Environmental
Problems as an Interactive Learning Process I
MPdeWit
Centre for Scientific and Industrial Research, Pretoria
ABSTRACT
The foremost limitation of public policy approaches is that the context of the
public policy problem is not taken into account. In the case of complex and
dynamic environmental problems, such as global climate change, there is a need
for a framework for approaching economic policy that takes account of the
complexity and changing realities of such problems. The objective of this paper
is to present a framework to approach economic policy making in a case of such
complex and dynamic environmental problems. The literature on economic and
public policy theories, the need for a systematic policy design process and
approaches to complexity and dynamics in policy making is framework
available to one where the focus is on the best learning process to facilitate
economic policy making on complex and dynamic environmental problems.
Based on sociological models of experiential learning, a multiple-loop learning
framework (MLLF) is presented. This model illustrates the importance of
orchestrated science-policy interactions through interactive learning. The
opportunities and limitations of this model are discussed with reference to the
debate on economic policy for global climate change.
JEL 020, 028, 038
1 INTRODUCTION
The sheer complexity and dynamics of global environmental problems, such as
climate change, has opened an intensive debate on the approaches of various
economic theories to environmental policy making. It is realized that climate
change is also an economic problem as scarce resources are either vulnerable to
climate change or have to be allocated to the mitigation of climate change
(Nordhaus, 1991; Cline, 1992; Fankhauser, 1995; Bruce, Lee & Haites, 1996;
IPCC, 200 I). The question, however, is whether various economic decision-
rules are necessary and sufficient to inform economic policy on global climate
change, and how this should be done.
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
SAJEMS NS Vol 5 (2002) No 2 309
The objective of this paper is to answer the question of how economic policy
making for complex and dynamic environmental problems can best be
approached. In section 2 the underlying positivistic economic philosophy and its
implications for approaching policy making is discussed with reference to
various economic theories on the environment. In section 3 it is emphasized that
a systematic policy design process is needed that can take account of complex
and dynamic policy problems. In section 4 the literature on existing approaches
to complexity and dynamics in policy making for sustainable development is
reviewed which sets the scene for the development of policy learning
framework in section 5. In section 6 a few words are said about current
economic policy making approaches to climate change and the need for the
application of an interactive learning process. Section 7 concludes.
2 FROM ECONOMIC THEORY ON THE ENVIRONMENT TO
POLICY MAKING
The flourishing economic-philosophical argument is that economics is a well-
defined science underpinned by realistic assumptions. This positive approach to
economic problems is defInitive and realism carulOt be an independent criterion
(see Hahn & Hollis, 1979: 2). The application of the positive economic approach
to policy making was outlined by Keynes (1917) in his book The scope and
method of political economy. He urged the importance of "recognising a distinct
positive science of political economy" (Friedman 1953: 3). The stage was set for
the further development of economic political theory and, therefore, subsequently
economic policy making approaches to issues pertaining to the natural
environment. To paraphrase Friedman (1953: 4):
Positive economics is in principle independent of any particular ethical
position or nonnative judgements. As Keynes says, it deals with "what is ",
not with "what ought to be ". Its task is to provide a system of
generalisations that can be used to make correct predictions about the
consequences of any change in circumstances. Its peifonnance is to be
judged by the precision, scope, and confonnity with experience of the
predictions it yields. In short, positive economics is, or can be, an
"objective" science, in precisely the same sense as any of the physical
sciences.
The link between economics and scientific objectivity has been made and
translated into the realm of economic approaches to policy through specific
welfaristic decision rules such as Pareto optimality. Such approaches have been
standard in approaching economic policy issues ever since.
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
310 SAJEMS NS Vol 5 (2002) No 2
Positivistic economics accepts the scientific objectivity (what is or what might be),
or the ethical neutrality, of economics (Bromley, 1991: 207). Normative
economics is perceived as connoting value-laden arguments (what ought to be).
According to Blaug (1992: 112) one can best think of this as a distinction between
scientific economics and practical advice on economic policy. Inspired by the
philosopher David Hume (1711-76), the perception of a watertight distinction
between the realm of facts and the realm of values was born and further developed
within positivistic economics.
The methodology of positive economics, however, has not remained unchallenged
in economic thinking. Several scholars have asked questions on the nature of and
the assumptions underlying positive economics (Ha1m & Hollis, 1979; Blaug,
1992, 1980; Boland, 1982; Caldwell, 1982; Eichner, 1983; Bromley, 1991, 2000).
Blaug (1992, 1980) and Caldwell (1982) come to the conclusion that economics
cannot claim to be scientifically objective and ethically neutral. Ward (1972)
argues ''that economics is basically a normative science adorning itself with the fig
leaf of hard-headed positivism" (as quoted in Blaug, 1992: 238). The distinction
between what is and what ought to be in such problems, cannot be separated easily.
Positivistic economics became divorced from value-judgements and too many
values were treated as basic or were simply ignored (Sen, 1970). There is no real
value-free social science. Economy is partly ideology, and a separation of the
positive from the normative in developing economic theory is impossible.
Approaching economic policy to complex and dynamic problems would involve a
rethink on Keynes' positive science of political economy.
The implications of such a statement on policy making are significant (Bromley,
1991: 212). The positivistic policy research programme in economics has
concentrated on an apparently value-free way of participating in the policy debate.
The resulting new welfare economics, with efficiency as evidence of scientific
objectivity, however, is controversial. Blaug (1980: 147-8) concluded that ''the
concept of Pareto optimality ... should not be confused with theorems of positive
economics .. .immense confusion has been sown on matters of 'efficiency' without
committing ourselves to any value judgements". The search in welfare economics
for a fixed order for the collective choice process is fiustrated by the complexity of
policy making. Bromley (1991: 217) describes this search for a fixed order as
reductionistic decision rules. According to Bromley (1991: 217), the persistent
debate on the appropriate welfare criterion and on the Boadway paradox, in which
the ability of the gainers to compensate the losers does not lead to an unambiguous
improvement in social welfare (Boadway, 1974), is testimony to this complexity.
Efficiency on the policy level is a value-laden concept in itself. This change in
economic thinking, especially on the level of policy making, forces a recognition
of the complexities in the collective choice process.
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
SAJEMS NS Vo15 (2002) No 2 311
Now that the apparent simplistic theory-policy relationship from positivistic
economic approaches has been questioned, the next question is what alternative
public policy approaches could add to the debate. Schneider and Ingram (1997: xi)
identify four prominent theories that have guided politics and policy during most
of the twentieth century: the pluralistic theory, policy sciences, public choice, and
critical theories. Schneider and Ingram (1997: 10-11) conclude that not one of
these public policy approaches "have provided an adequate diagnosis, explanation.
or prescription for public policy and that their central failures include a narrow
knowledge orientation, insufficient attention to policy design, and a failure to
recognise the central role social constructions have in determining the
characteristics of policy designs". The foremost limitation of these policy
approaches is the assumption that the context of the public policy problem is not
taken into account, i.e. one design fits all, thereby creating an imbalance in the
other values policy must serve (Schneider & Ingram 1997: 10).
The point that public policy, and more specifically environmental policy, is
context-specific has also been pointed out by other scholars. According to
Gunningham and Grabosky (1998: 32): "So complex and various are the causes of
environmental degradation and the circumstances in which they arise that no single
instrument, and indeed no single mix of instruments, could conceivably be
successful in addressing all or even most of them. The optimal policy could only
be determined on a case-by-case basis". Opschoor and Turner (1994) state that a
priori rules are inferior to case-by-case analysis. Sterner (1994: 7) argues that the
choice of "[a] policy instrument ... depends crucially on both the institutional
setting and on the exact nature of the environmental problem at hand".
Nevertheless, most economic theories on the environment link most closely with
the public choice approach to public policy. In't Veld and Kraan (1991) present a
useful overview of different public choice models, distinguishing between radical
public choice, individual moralism, constitutional economics and a hybrid of these
three, namely, moderate public choice:
In the radical public cboice model, the neoclassical model with a homo
economicus is extended to the political process without any adjustments for
the realities of the political process. Brennan and Buchanan (1985: xi)
describe this model as one attempting to offer a pure science of politics that
is fully analogous to the science of markets. According to them the objective
is "to derive testable hypotheses about the effects of specific changes in
basic parameters on observed political results".
In the model of individual moralism political behaviour is predominantly
determined by moral motives, in contrast to the economic behaviour of homo
economicus. The task of political theory is to clarify the choices people are
making in terms of their belief systems and to improve the contents of these
belief systems themselves (In't Veld & Kraan 1991: 3). Empirical work is
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
312 SAJEMS NS Vo15 (2002) No 2
directed at description and analysis of prevailing political belief systems.
Brennan and Buchanan (1985: ix) describe this model as being concerned
with "analytical esoterica in the modelling of alternative political
arrangements". Modern welfare theory with its ernphasis on the
development of the social welfare function determined by normative criteria
such as economic efficiency or sustainability, is an example of this public
choice model.
The basic claim in the model as developed by Buchanan (1991, 1987),
namely constitutional economics, is that theoretical analysis should proceed
in different ways for different kinds of political decisions. The fundamental
distinction is between decisions at the constitutional level of choice and the
post-constitutional level of choice. The first type of decision determines the
legal-institutiona1-constitutional structure of the polity, while the second
type of decision detennines the economic and political process within a
given structure (Buchanan, 1987). In post-constitutional decision making the
model of homo economicus applies. At the level of constitutional choice
(choice about institutional rules) decision makers are susceptible to what In't
Veld and Kraan (1991: 4) label normative reasoning. However, individual
choice remains dependent on the institutional-constitutional constraints
(Buchanan, 1989). The overall objective is "to understand the workings of
alternative political institutions so that choices among such institutions (or
structures of rules) can be more fully infonned" (Brennan & Buchanan
1985: xi).
The model of moderate public cboice is closer to the radical public choice
model in that it appreciates the relevance of positive economic analysis for a
broader area of economic decision making than constitutional economics
tend to do (In't Veld & Kraan 1991: 5). On the other hand, individual
moralism is more accepted in that there is some room for normative
reasoning at the post-constitutional level of choice. In some areas of post-
constitutional choice, moral motives can be decisive for the behaviour of
citizens, politicians and bureaucrats. This model builds on the model of
constitutional economics, but pleads for a more realistic mix of positive
analysis and nonnative reasoning.
The next question is how these public policy theories relate to economic theories
on economic-environment interactions. It can be expected that the different
economic theories on the environment have different inclinations towards a certain,
or more than one, public policy modeL It is important to link economic theories on
the environment to public policy models so as to lose no useful insights into policy
design through the ex ante choice of a particular public policy paradigm. In Table I
an attempt is made to link economic theories on the environment to public policy
models. This is an early attempt, as most economic theories on the environment,
especially ecological economics, are in early phases of development.
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
SAJEMS NS Vol 5 (2002) No 2 313
Table 1 Economic theories on the environment and public policy
models
Public choice Pluralism I Po~tical Critical
! sCience theories
Radical Individual: Constitu-
moralism tional
economics i
Environmental
X.
Economics
x.
i
Ecological
X. x. x.
Economics
Neo-institu-
tional x. x. X.
economics
Evolutionary
X?
I approaches i
Source: De Wit (2001: 114)
The' focus on optimisation within a particular setting and according to a set of
pre-determined normative criteria, such as economic efficiency (in the case of
environmental economics) or sustainability (in the case of ecological
economics), excludes an unconditional acceptance of the status quo (pluralism)
or a total rejection of the status quo (critical theory).
The evolutionary approach, however, comes closest to accepting processional
realities2 as a foundation for theoretical development, but the reviewed literature
does not support the uncritical acceptance of the status quo. The lack of focus
on optimisation rules out any connection with public choice models and
political science. It can be hypothesised that some elements of critical theory,
such as the emphasis on social change, are aligned with evolutionary
approaches, but no conclusions can be made at this stage.
Both environmental economic and neo-institutional approaches are extensions
of positivistic neoclassical economic theory. In this approach the theory of
homo economicus could be extended to the public policy arena, as in the radical
public choice approach. Some free marketeers within the environmental
economic and neo-institutional approaches follow this approach (Anderson &
Leal, 1991; Anderson & Hill, 1995). This is the exception rather than the rule,
as most environmental economic scholars follow assumptions based on (new)
welfare economics. Through their emphasis on economic efficiency, these
theories are more inclined to follow the welfaristic approach of individual
moralism. The difference is that neo-institutional approaches are closely aligned
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
314 SAJEMS NS VolS (2002) No 2
with constitutional economics, while environmental economics is not. In
constitutional economic theory the co-existence of political and market
institutions are accepted by. defmition, which aligns closely to the neo-
institutional notion of a variety in property right structures. None of these
institutions is left in its black box, as environmental economists tend to do with
the political process (see Hahn (1989) for this critique on environmental
economic policy making). No literature was found to support the alignment of
environmental economics or neo-institutional economics with one of the non-
traditional economic approaches to public policy. In summary, environmental
economists are clearly linked to public policy by individual moralism, while
neo-institutional economists align more closely with a mix of constitutional
economics and individual moralism, but with an emphasis on the former.
Ecological economic approaches take a cautious approach to environmental
degradation, distrusting in general the ability of markets to provide a sustainable
solution. Although a better defined role for both markets and governments is
accepted when compared to the environmental economic approach, the realities
of the political process itself are not effectively internalised. Daly (1992, 1999)
suggests a logical sequence of policy tasks: defining scale and distribution in a
political process and leaving the allocation of resources, within these
constraints, to the market. Peet (1992: 220-21) refers to this sequence as a
policy hierarchy, since, in his view, "ethical and ecological principles are the
true determinants of price". Cumberland (1994) brings the concepts of interest
group acceptability and political feasibility into the debate and argues that the
policy approach depends on the ecological damages in the frrst place. In the
case of non-measurable damages, the property-rights approach is sufficient.
When damages are measurable, but do not lower the ecosystem productivity, an
incentive-based approach (through economic instruments) is applicable.
However, when non-sustainable long-term damages are expected to occur, a
regulatory approach is followed. For a graphic illustration of this approach, see
Figure 1.
Ecological economists follow the precautionary approach arguing a priori that
long-term damages are non-sustainable. The property rights and incentive-based
approaches can be useful, but only within the boundaries of the politically pre-
defined scale and distributional constraints. The social welfare function, as
identified in welfare theory, is still maximised, but subject to different
constraints as in environmental economic and neo-institutional economic
approaches. Ecological economic approaches are therefore closely linked to
individual moralism, but this conclusion is not sufficient. A clear role for
government intervention is identified, thereby excluding any links with the
radical public choice approach.
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
SAJEMS NS Vol 5 (2002) No 2
Figure 1 Ecological damage and policy
.r:.C')IO~;tCOJ damage
Marginal Damage
Non-sustainable
long-term damage
~----------~------------~
Measurable
damage
decreasing
productivity
Non-measurable
damage
Property right
abatement
Source: Cumberland (1994)
Incentive
315
Although the links have not been made explicitly in the literature, the sequential
approach to policy making as promoted by ecological economists such as Daly
(1999), is similar to the constitutional economic approach of constitutional and
post-constitutional choice (Buchanan, 1991). Politically normative choices on
scale and distribution are made on the constitutional level, while the market can
allocate resources efficiently within this set of pre-defined constraints_ These
approaches also link with the policy science approach of focusing on science
and professionalised bureaucracies in improving policy design. An example is
the emphasis on the scientific derivation of ecological indicators needed for
standard-setting. The general lack of attention on the power relations between
different actors, and the desire to change the status quo towards a more
sustainable society excludes the pluralistic approach to public policy>. In
summary, the ecological economic approach links best to public policy models
through individual moralism, constitutional economics and political science. It
can also be argued that the strongest link is with individual moralism from
welfare theory, but within the boundaries set by constitutional economic and
policy science processes.
One could debate on the relative contributions of various public policy approaches
to the various economic theories on the environment, but for the purpose of this
paper it is clear that a focus on one specific economic theory and one specific
public policy approach leads to a partial interpretation of reality, resulting in tum in
a bias, and possibly increased complexity, in policy recommendations. The
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
316 SAJEMS NS Vol 5 (2002) No 2
argument also exposes the fact that an ex ante implicit choice of a public policy
approach needs to be made, and this reinforces the need for more transparent
policy design.
3 THE NEED FOR A SYSTEMATIC POLICY DESIGN PROCESS
The public policy approaches discussed so far all view the policy maker as
exogenous to a policy support system. Following the work of March and Simon
(1958) and Simon (1977), it is assumed that policy making follows a linear and
one-dimensional three-stage process of intelligence, design and choice (see
Checkland & Scholes, 1990: 165). In the first stage, problems are identiffed and
data collected; the design-phase consists of plaruring for possible alternative
solutions; and the choice-phase entails the selection of an alternative and the
monitoring of its implementation. Based on 12 case histories of policy reform in
developing countries, Thomas and Grindle (1990: 1165) refer to a similar linear
process in the implementation of policy reforms, following the stages of agenda
setting, decision and implementation. Hirschman and Lindblom (1972: 358) also
critically refer to the linear process of rational decision making in policy making
where objectives or values are first clarified, secondly, alternative means of
reaching these objectives are surveyed, thirdly, the consequences of each
altemative are identified, and fourthly, each set of consequences is evaluated in the
light of the objectives.
The need for an alternative approach to this linear process of policy making has
been raised by various scholars (Lindblom, 1959, 1968, 1979; Hirschman &
Lindblom, 1972; Thomas & Grindle, 1990; De Greene, 1993). Lindblom (1959)
argued against proposals for decision-making tools without an explicit recognition
of the realities of the political process itself. This point of departure closely
corresponds to the argument that the analysis of the political process itself has to be
included in a systems approach to policy making. This approach is, in fact, a
recognition of the dynamic, changing nexus of decision-making issues and
cognitive issues. As pointed out by Hirschman and Lindblom (1972: 357), the
critique centres around two assumptions implicit in most of the linear approaches
to policy making:
Firstly, that public policy issues can best be understood by attempting to
understand them. What does that mean? They cannot be theorised!
generalised?
Secondly, that there exists sufficient agreement to provide adequate criteria
for choosing among possible alternative policies.
On the first assumption, the public policy making process should be viewed as part
of a system characterised by continuous feedback. Hirschman and Lindblom were
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
SAJEMS NS Vol 5 (2002) No 2 317
writing long before the scientific revolution of dynamic, non-linear systems took
place\ but the results only seem to confirm their original thesis. It is argued in the
literature on dynamic non-linear systems theory that unpredictability and novelty
will always be around us, even in the case of deterministic complex systems
(Stewart, 1990). None of the public policy approaches, as discussed earlier,
referred explicitly to the dynamic complexity of the policy making process. As
discussed, the foremost limitation of all the public policies evaluated is the
assumption that the context of the public policy problem is not taken into account,
i.e. one design fits all. Unpredictability and novelty do not feature meaningfully in
these approaches. The recognition, however, that the policy process itself is
continuously changing, emphasises the limitations of the perception that the
systems approach can only be seen as a decision support tool to a relatively fixed
policy process. The policy process co-evolves with the problem situation (De
Greene, 1993: 7) and is part of a dynamically interacting systems-field (De Greene,
1993: 133).
The policy process has also long been debated in the development literature.
Quarles van Ufford (1993), for example, criticised the lack of attention to the
process of policy formulation itself. His criticism focuses on an optimised systems
approach, in effect, a closed political system - one where development is presented
as manageable, intellectually simple and explicable to an ignorant audience (Van
Ufford. 1993: 157). His call for a model, that is a better balance between normative
(or ideological) and analytical models, is what is understood in an open, systems
approach to the policy making process one that explicitly recognises the
importance of feedback and continuous learning.
On the second point, the linear process of policy making can only take place on the
basis that sufficient agreement exists on adequate criteria for choosing among
policy alternatives (Hirschman & Lindblom, 1972: 357). This approach does not
advocate either laissez faire or expanded political influence, but is concerned with
decision-making and problem-solving activities carried out by the political
authorities (Hirschman & Lindblom, 1972: 364-365). It was already suggested that
such agreement is often unlikely as many public policy approaches and different
value principles exist with very different implications for policy design, such as
economic efficiency and sustainability norms.
In view of the above, the design of economic policy for complex and dynamic
problems cannot simply be based on theories on the interface between economics
and the environment. There are many different theories, such as environmental
economics, ecological economics, nec-institutional economics and evolutionary
economics, that have different frameworks for the design of policy. None of these
theories can make a claim to represent absolute truth values; all of them have to
intemalise increasing spatio-temporal scales of environmental problems and in one
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
318 SAmMS NS Vol 5 (2002) No 2
way or another, both the static and dynamic aspects of political reality need to be
taken into account. Public policy approaches are also divided in their approach and
based on ideological a priori first principles that are often closely linked to
economic or other theories (Schneider & Ingram, 1997). The design of economic
approaches to policy regarding complex and dynamic problems needs to be
approached as a process that incorporates feedback and is geared towards
continuous learning.
It has been pointed out that the systems approach to sustainable development can
take account of both increased complexity and the dynamic processes of change
(Clark, Perez-Trejo & Allen, 1995; De Wit, 2001). The question is whether such
an approach can be used for policy making as well. The added value of this
approach is that dynamic systems approaches are not only used to enhance the
presentation of specific problems to policy makers, but to include interest groups in
a learning process of policy making. The design of policy is not confined to
analysts, but a far broader approach is taken to both the sources of knowledge and
participation. There is thus a need for a policy process that intemalises different
values within their context, before making a choice on a policy framework. Such a
process has to be able to include the many aspects of complexity, but also the
dynamic character of policy making for sustainable development. The
development of such a framework can best be informed by existing literature on
the interface between policy making and complexity theory.
4 APPROACHES TO COMPLEXITY AND DYNAMICS IN POLICY
MAKING FOR SUSTAINABLE DEVELOPMENT
The search now extends to identifying approaches to policy making that explicitly
takes account of complexity and dynamics in real world environmental problems.
The question that needs to be answered is how the complexity in economy-
environment-policy systems and the changing realities of policy design can best be
approached.
The application of dynamic systems approaches to policy making processes can be
informed by various contributions in the literature on systems dynamics,
management and organisation (Gill, 1996; Saeed, 1994; De Greene, 1993; Rastogi,
1992; Gardiner & Ford, 1980; Murthy, 2000; Warfield, 1999, 1994; Vennix, 1999,
Checkland & Scholes, 1990; Checkland, 1981; Senge, 1990; Stacey, 1995;
Kelleher, 1970; Busterud, 1977l In addition, and the focus of this paper, a
relatively small amount of literature exists on the application of the dynamic
systems approach to policy making for environmental problems and/or sustainable
development (Saeed, 1998; Roe, 1998; Comfort, 1999; Schoot-Uiterkamp, 1999;
Schmoldt & Peterson, 1999; Wolfenden, 2000; UNDP, 1999).
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
SAJEMS NS Vol 5 (2002) No 2 319
Saeed (1998) argues that informing policy making for sustainable development is,
given its high complexity and uncertainty, an experientialleaming process. The
complexity of sustainable development is captured in a system dynamics model
that informs policy makers on appropriate actions. Such a systems dynamics model
is dependent on the learning process followed by the modeller.
Figure 2 Kolb's model of experiential learning
Concrete
Experience
Testing __ A_CtI_'v_e ___ Pas_s_iv_e -I_I Watching I Observations
ImplicatiOllS '---.----' . • . and reflecllOllS
:, .. J
Abstract
Conceptualisation
Source: Kolb (1984) as quoted in Saeed (1998)
The core competencies of experiential learning are the learning faculties of
watching, thinking, doing and feeling in Kolb's model (see Figure 2) (Saeed, 1998:
407). An important characteristic of these tasks is that they correspond to passive-
physical, concrete-cognitive, active-physical and abstract-cognitive domains. Such
a display takes account of two basic human functions, namely the physical and the
cognitive, and integration along two primary dimensions, namely the passive-
active and concrete-abstract dimensions. The important conclusion from such an
approach is that an understanding of complexity can be facilitated through a
learning model that takes various dimensions of human functioning into account
Saeed (1998: 407,409) likens system dynamics modelling to an art of balancing
between these learning faculties.
Roe (1998) also argues that complexity has to be taken seriously when formulating
policy, but unlike Saeed (1998), he does not take a system dynamics modelling
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
320 SAJEMS NS Vol 5 (2002) No 2
approach. A central problem of public policy analysis has been to fmd new ways of
analysing issues of increasing uncertainty and complexity. Roe suggests the
method of triangulation - using various methods, databases, theories and
approaches to converge on the complex issue in question. In this method the same
set of questions regarding policy issues are asked from multiple perspectives. In
this way facets of the problem are identified that may be omitted by other methods.
Comfort (1999: 182) welcomes this approach in that it attempts to find appropriate
methods that fit the problems, rather than focusing on problems that fit the existing
methods. Roe applies this method to the issue of sustainable development and
makes recommendations on the basis of selected perspectives on the issue. Without
discussing the recommendations in detail, the important fmding for the pwpose of
this paper is, as Comfort (1999: 183) puts it: " ... [Roe] puts to rest old arguments of
politics versus economics in policy debates and demonstrates afresh that both are
central elements in designing workable policy, and neither are likely to define the
policy issue accurately". And further (1999: 183): ..... implicit in his approach is
the recognition that invalid policy analysis contributes to the clamour and
confusion in the policy debate. Rather than reducing complexity and increasing
clarity of understanding on problems to facilitate action, methods of analysis that
provide inaccurate or partial views of a complex policy issue seriously inhibit both
understanding and action".
The United Nations Development Programme (UNDP, 1999) used Roe's method
to formulate policy for sustainable livelihoods by starting to recognise that such
systems are complex and adaptive and can best be analysed on a landscape level.
The UNDP (1999: 4) distinguished between two types of policy analysis, namely
conventional and complex. Conventional policy analysis is sequential and
cumulative; complex policy analysis is interactive and convergent. The former is
only applicable in situations "oflow environmental uncertainty, relative stability in
public (including government) objectives, strong institutional memory in the
design and implementation agencies, and sufficient resources to tolerate mistakes
in trial and error learning relevant to the issue in question" (UNDP, 1999: 4).
Complex policy analysis is further applied to the issue of sustainable livelihoods,
using the method of triangulation as discussed in the previous paragraph. The main
conclusions are that complex policy issues demand analysis on a case-by-case
basis.
These approaches do take account of complexity in policy design and attempt to
deal with this through systems modelling and triangulation approaches. This is a
step in the right direction, but the limitation with these approaches, and in fact,
with most of the other approaches to understanding complexity (see De Wit, 2001:
143-53), is that the policy maker is treated exogenous to the understanding of the
policy problem. The focus is on the analyst or scientist to understand and make
recommendations on a better understanding of the complexity in the problems at
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
SAJEMS NS Vol 5 (2002) No 2 321
hand. Although a huge body of literature recognises the need for consultative,
participatory or group learning processes, the overall impression is that these only
inform policy making, and does not really endogenise the policy design process
within the system under evaluation. However, a limited body of literature is
available that explicitly attempts to provide a model to handle both complex issues
and design a process wherein various actors (such as policy makers themselves)
can be involved.
Checkland and Scholes (1990) apply a soft systems methodology (SSM) trying to
'manage' real-world situations; it is system thinking based and is applicable to
taking purposeful action to change real situations constructively. In the SSM it is
acknowledged that problem solving can be much more complex and dynamic than
allowed for in the more traditional linear approach. Checkland and Scholes (1990:
280-84) distinguish between two modes of SSM. In mode I the key word is
intervention. SSM is used to do a particular study - it therefore serves as a
decision tool for improved management of complex systems. In mode 2, and the
one we are interested in, the key word is interaction. The work is done through the
use of SSM - real-world feedback is included in a management model of
continuous learning. 'This does not mean continuous learning about the problem
situation per definition, but continuous learning about how to deal with these (new)
streams of information in the management process itself. In a model where
decision makers are endogenous to the model, mode 2 SSM is more applicable.
The important point is that intervention analysis, social system analysis and
political system analysis, which are all aspects of the SSM model, inform an
understanding of the problem situation and root definitions of the system6• This
means that the roles, norms and values of participants are included in the analysis
of the system and are subject to continuous learning and revision. The conceptual
models are built on the basis of an initial understanding of the system and
compared with perceived reality through informal discussions, formal questioning,
scenario writing based on operating the models, and attempts to model the real
world in the same structure as these conceptual models (Checkland, 1981). If there
are major differences, this feedback is internalised in a second round of enquiry.
The resulting changes that are proposed are subsequently tested for systematically
desirable and culturally feasible changes. Based on experiences with SSM, such a
process can best be attempted through participation and interaction between
problem solvers and actors in the system, and not only by a team of independent
analysts. In summary, learning about complex systems is facilitated through
streams of enquiry and the need for dynamic feedback is recognised.
In the same vein, Vennix (1999: 391) further argues that an effective way to
address messy problems would be to remove barriers to learning through
participation in the construction of the model using methods such as group model
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
322 SAJEMS NS Vol 5 (2002) No 2
building. A messy problem is defined as a situation in which opinions in a
management team differ considerably (polarisation) (Vennix, 1999: 370). Vennix
emphasises the importance of team learning and effective communication to
overcome cognitive barriers and accept differences in perceptions that lead to
multiple realities. Such team learning would need an effective facilitation. The
important contribution for this paper is that the process of decision making in
messy problems needs to be facilitated carefully through group model building
techniques. There are barriers to cognitive learning and communication that would
impede the formulation of solutions for these problems.
5 THE MULTIPLE LOOP LEARNING FRAMEWORK (MLLF)
The challenge remains to develop a framework for economic policy on complex
and dynamic environmental problems. It is important to think about the meaning of
the word framework as this can be interpreted in many different ways. Pollard and
Liebeck (1994: 316) defmes a framework as "the structural basis of an
organisation, the structure of a plan, etc". One can therefore describe a framework
as something that approximates the best possible solution as closely as possible.
The question: What is best?, however, implies reference to pre-determined criteria
outside the framework itself (e.g. economic efficiency, equity, sustainability or
survival). The discussion of such a framework would either take one or more of
these criteria as the point-of-departure, or fall into a discussion of the relative
weights of the different criteria. Such an approach, however, would defy the
argument that uncertainty and novelty are features of complex and dynamic
systems, and that learning about these systems are the only available option. In
such a case the question on a framework for economic policy making for complex
and dynamic environmental problems is rather different. The focus changes from
finding the best framework available in the circumstances to one where the focus is
on the best learning process to facilitate economic policy making. What aspects
have to be in place and what questions have to be asked and at what time/stage to
guide the process of learning on economic policy making for complex and
dynamic environmental problems?
This brings us to the next question: How does one learn? The answer to this
question would delimitate some of the issues that need to be included in a learning
framework. Kolb's model, as discussed in the last section (and referred to in Figure
2), is a useful starting point in this regard. One learns both cognitively and
physically, but in a circle of continuous feedback.
Do policy makers also learn in such a way? And, ifnot, why not? Dror's (1971)
reference to a Scbool of Rulers accentuates the fact that such questions have been
raised before. There is no reason to believe that policy makers themselves are not
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
SAJEMS NS Vol 5 (2002) No 2 323
in a continuous cycle of learning, or should be in such a learning process, given the
uncertainty and complexity pertaining to policy making issues on complex and
dynamic environmental problems. However, the resulting flexible policy making
approach could best be guided by a process of learning, in order to reach the best
learning process possible.
Are there situations when the learning process is finalised? The point raised that
complexity and novelty will always be around us, suggests that such a process
cannot ever be finalised. However, it could be argued that the intensity of learning
would differ greatly depending on the problem at hand. Based on Kolb's model
(Figure 2), one could define learning on two levels:
cognitively focused learning cycle (vertical axis)
practically focused learning cycle (horizontal axis).
In the first instance, the added value of analytical work, such as applications of
economic theories on the environment would prove to be great. The application
of economic analysis on environmental problems would provide important
inputs to the policy making process. In case of a practically focused learning
cycle, the marginal added value of cognitive contributions is declining and
added value to the policy making process is gained through practical
implementation of policy and attempting to recognise patterns in the responding
economies. In the problem at hand, and following the distinction between
endogenous and exogenous policy making, one could distinguish between a
learning framework for policy makers and a learning process for analysts and
system dynamic modellers. The latter is a process focusing on the cognitive
content of learning and the former focusing on the practical content of learning.
In a sense this is following from the contribution from Checkland and Scholes
(1990) who argue that a mode 2 interactive approach is required to deal with the
process of policy design itself. Within this meta-level approach one could defme
a problem-solving approach that takes more cognisance of case-specific
analytical issues, as discussed in Checkland & Scholes's mode 1 approach.
The concept of a learning framework would include a model that facilitates
learning itself on different levels. One can refer to the existence of learning
cycles within learning cycles, analytical learning cycles within policy process
learning cycles, or multiple loop learning cycles, to represent the changing
emphasis on cognitive and physical aspects of different learning events for both
analysts and policy makers. The issues raised can be clarified in a simple
conceptual framework that captures the most important issues in a framework
for economic policy in case of complex and dynamic envirol1mental problems.
Such a multiple loop learning framework (MLLF) is illustrated in Figure 2.
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
324 SAJEMS NS Vol 5 (2002) No 2
Figure 2 A multiple-loop learning framework (MLLF)
~+-~---E--+----'3-----i--'--------jiM Watching
'-----'
4
o
Source: De Wit (2001: 160)
The key research question how best to approach economic policy making for
complex and dynamic environmental problems can be answered on a conceptual
level with the aid of a MLLF. The MLLF illustrates a number of pertinent
issues:
Economic policy making for complex and dynamic environmental problems
is first and foremost a learning process, comprising the learning faculties of
watching, thinking, doing and feeling.
This learning process has to be facilitated in the best way possible, and the
key learning faculties balanced in the right way.
This balance would differ with the problem at hand. In loop I (ABeD), the
policy process loop, the focus would be more on the physical and more
practical aspects of group processes and facilitation, while loop n (1234), the
theoretical learning loop, would be more cognitive, focusing on applications
of economic theories on the environment and public policy theories.
These loops, however, are not separate but continuously influence each other
when policy making is endogenous to problem resolution. Without
elaborating on the interaction between theoretical learning and policy
process loops, which is a process of well-facilitated group participation, the
important point is that both these loops have to be facilitated for optimal
cross-pollination.
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
SAJEMS NS Vol 5 (2002) No 2 325
The nearer to the epicentre (the diamond in the centre), the higher the
applicability of high precision science. Closer to the epicentre of the model
is the only area where continuous learning is not required, because of low
complexity and low randomness. In such cases the problem is too simple to
establish a process of policy and analytical learning.
In situations where the problem under observation is too unorganised to
attempt a process of learning, short-cut averages could be used to indicate
the macro direction of the system, but at the expense of learning about the
key levers of control, change and constraints.
This conceptual framework illustrates a range of key concepts when dealing
with complex and dynamic environmental problems. Such types of problems
would require continuous learning systems, on both the theoretical and political
level and in interaction with each other. In cases of complex and dynamic
problems, theoretical studies should be infonned by political variables as well
and political learning processes should be facilitated as much as possible to
optimise on available theoretical infonnation.
A few pertinent questions on especially the economic approach to
environmental policy remain. What does the MLLF mean for the economic
approach to environmental policy? What are the key learning points of the
suggested policy learning framework for economic theories on policy making?
A few issues are highlighted:
To base economic approaches to policy making on positive scientific
approaches alone impedes the ability to reach the best solutions for
complex and dynamic problems. Economic theories on the environment
are by definition not scientifically objective. They are based on various
different normative premises.
The nonnative points of departure of various economic theories on the
environment are critically examined in an interactive policy learning
framework. Economic approaches to environmental policy making are
partial representations of reality. The results of such studies are placed in
a facilitation process and allocated to the type of problem at hand. Studies
that, for instance, demonstrate the relative costs and benefits of alternative
approaches to environmental policy making have the highest value in
settings where uncertainty and polarisation of environmental problems are
minimised.
Solutions to complex and dynamic environmental problems should not be
sought in the linear internalisation of external effects alone, but supported
and infonned by a process of interactive learning and feedback. The focus
on the first instance, shifts from internalisation of external effects inside
the system to the internalisation of the policy making process itself in
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
326 SAJEMS NS Vol 5 (2002) No 2
addressing complex and dynamic environmental problems. Economics
and politics are complementary elements in the design of policy for
complex and dynamic environmental problems.
The opposing policy advices from various theories on economy-
environment interactions are counterproductive. These theories depart
from implicit pre-analytical frameworks that do not represent any notion
of truth-values by definition. Rather than building theoretical power
blocks on the basis of different economic approaches to environmental
policy, the challenge is to develop and apply the most relevant approaches
in the right problem context. In effect, it is a move away from a universal
fit of a particular theory on economy-environment interactions to one of
contextual design.
Economic approaches to environmental policy can be leveraged to their
highest use when integrated in a well-designed group facilitation process.
These processes should at least intemalise the following issues
7
:
- the normative points of departure of interest groups and different
theories (partial interpretation of reality)
the wide range of scientific, economic and political theories on
addressing such a problem (type of uncertainty)
the different interest groups and their values (level of polarisation)
the spatio-temporal scale (context) of the problem at hand and the best
theoretical fit to these problems (type of complexity).
In short, the MLLF does not exclude economic approaches to environmental
policy, but is a meta-level process that facilitates the application of such
economic approaches to their best use. The important learning points for policy
making on complex and dynamic environment problems are that policy makers
have to recognise that analysis of the problem will be a learning process unless
the type of problem is too simple to justify learning. The policy makers' choice
is:
to slice complex and dynamic environmental problems into simple parts,
but at the same time have a well designed policy learning process in place
to internalise and facilitate partial analytical results. This choice will be
biased towards certain normative criteria (e.g. economic efficiency,
sustainability) employed within analytical recommendations, but these
are at least placed within the context of a broader model of policy
learning.
to have an interactive multiple-loop policy learning framework in place
that optimises on learning from the cognitively focused analytical
learning cycle and the practically focused policy learning cycle.
The choice of a policy design process for complex and dynamic environmental
problems will have to be measured against the relative transactional costs of
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
SAmMS NS Vol 5 (2002) No 2 327
such a process. Although the benefits of an interactive learning process, as
illustrated by the MLLF, has been pointed out, the further practical question
remains what the costs of alternative systems will be. This is a question for
further research.
6 OPPORTUNITIES AND LIMITATIONS OF POLICY DESIGN
PROCESS IN THE CASE OF CLIMATE CHANGE
The direction of economic policy for global climate change is largely
summarised in publications from the IPCC, especially those of the working
groups on mitigation options (Bruce, Lee & Haites, 1996; IPCC, 2001). At the
time of writing the latest IPCC Working Group III (WG III) contribution to the
Third Assessment Report has been accepted, but not approved in detail.
Nevertheless, the IPCC's line of thought in this document on economic policy
issues could be taken as starting point for an evaluation.
The evaluation is focused on the issue of policy making and decision making on
climate change. The pertinent question used as point-of-departure in this review
is: What framework is used for approaching mitigation policy on climate
change? This question would inevitably produce an answer that could be
compared with the MLLF approach, on a conceptual level, as proposed in the
previous section.
The IPCC WG Ill's terms of reference have been to assess the scientific,
technical, environmental, economic and social aspects of the mitigation of
climate change. Although the mandate has been expanded from a disciplinary
assessment of the economic and social dimensions of climate change in the
Second Assessment Report (SAR) to an interdisciplinary assessment of the
mitigation options in the Third Assessment Report (TAR), this excludes explicit
reference to interaction between cognitive analysis of mitigation options and
practical policy making processes. The IPCC is very careful to distinguish that
it does not make any policy, but only assesses research that is 'policy-relevant'.
Although research is a political process, the idea is that policy decisions are
made in the UNFCCC COP process, and not by the IPCC8•
However, despite possible arguments to broaden the mandate, some useful
information in answering the question which approach to economic policy
making has been followed, could be found in IPCC WG III material. The
following issues are evaluated more specifically:
the nature of the mitigation challenge
the implementation of mitigation options
supportive decision making
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
328 SAJEMS NS Vol 5 (2002) No 2
6.1 The nature of the mitigation cbaUenge
Following the second assessment report (SAR) (Bruce, Lee & Haites, 1996), the
next !pee WG UI has received the mandate to include equity and sustainability
concerns in the assessment. This is a better representation of complexity in the
evaluation framework used. However, such an endogenous approach increases
the complexity of the system under observation. It has been emphasised in this
paper that learning about such complex systems has to be facilitated. This
means a well-balanced approach between the practical process of learning and
the cognitive content oflearning. However, the IPee WG UI line of thinking is
to redefine the nature of the mitigation challenge in terms of its complexity
focusing mainly on the content of the problem, but the process of achieving
such learning is not elaborated on in detail. In short, the focus in dynamic
complex problems is changing from an understanding of the nature of the
problem to support policy making towards an interaction with policy makers
and other actors in a dynamic policy making process. It is recommended that
economic policy makers on climate change evaluate the applicability of results
in the science of complexity and its application to policy making processes.
The approach to uncertainty and irreversibility taken in IPee WG UI work
would largely determine whether one would accept decision analytical tools or
more processional approaches to inform mitigation policies for climate change.
In any situation where time is treated as asymmetrical, one can talk about some
type of implied irreversibility. Whether future events are perceived to be
predictable depends on this type of irreversibility. Unpredictable processes are
those where novelty is emerging over time or where there is an unknown end to
future activities. The question relevant for the !pee WG UI work is which type
of irreversibility should be used for decision analysis of various problems
related to mitigation policy. An analysis of some issues at certain spatio-
temporal scales might be too uncertain to employ probabilistic outcomes. In
such cases, it would be an option to acknowledge complexity and manage the
problem-solving process accordingly. When the design of mitigative policies is
formulated as an open-ended., soft system, an interactive learning process can
provide various inputs for problem solving in the face of uncertainty 'and
irreversibility.
6.2 The implementation of mitigation options
The assessment of alternative policy instruments in terms of specific criteria is
an open-ended debate. The design of optimal policy mixes is, in practice, often
still static in orientation and dependent on the chosen normative criteria, such as
economic efficiency or sustainability. Despite a recognition in !pee WG UI
work of the importance of interest groups and policy makers themselves in the
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
SAJEMS NS Vol 5 (2002) No 2 329
design of policy instrwnents, an analysis of the policy design process that
underlies the selection of criteria and the relative weights attached to these
criteria in a learning framework, is an omission in this work so far. The
increased complexity of policy design for climate change (number of policy
instruments, number of criteria, number of actors) raises questions on the
analysis of such processes. It is recommended that the option of viewing the
problem as an interactive learning process that needs to be facilitated in the best
way possible, be evaluated.
6.3 Supportive decision making
There is no clear, unified, philosophical, theoretical or practical framework that
could guide policy making in all aspects of policy design. Various decision
support tools do exist, but neither on its own would provide a comprehensive
overview and recommended action on policies for climate change. IPCC WG III
acknowledges this and makes a sensible, reserved judgement on the usefulness
of these tools. However, the challenge remains to develop a systematic policy
design process that utilises these tools in the right context answering the right
questions at the right time. Again, the right balance between the cognitive
content of decision-making tools and the practical process wherein these tools
are utilised, is one worth exploring.
The discussion on decision-making frameworks is limited to providing advice to
policy makers that excludes interactive learning processes between cognitive
analysis and practical policy making processes. At best, the process of
international agreements is analysed in terms of the design of a framework for
institutions or to look at procedures of decision making at various levels.
Although these are important contributions in their own right, a facilitated
learning process that makes the most of cognitive inputs and provides inputs for
further cognitive development, is an omission in the focus on an analysis of
policy options as such. It is recommended that the focus of research on process
analysis is broadened from an analysis on the links between policy options and
the structure of agreements to one where the parameters for best cognitive-
practical interaction are spelt out and implemented. An additional research
question would therefore be: How can one best facilitate the interactive process
oflearning on mitigation policies for climate change?
7 CONCLUSIONS
In this paper it was attempted to formulate a conceptual framework for
economic policy making on complex and dynamic environmental problems.
Such a framework has been proposed in the form of a multiple-loop learning
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
330 SAlEMS NS Vol 5 (2002) No 2
framework (MLLF). The multiple loops consist of a practical policy making
leaming framework, and a theoretical analytical learning framework. These
leaming processes influence each other. The processes of learning on both
levels need to be facilitated and informed by tools such as group model building
and systems dynamics models.
An attempt to understand a problem is limited to situations of low complexity
and low randomness, such as non-living, engineering systems, consumer
choices between a small number of products and the growth of a plant under
controlled conditions. An attempt to ex ante understand a dynamic and complex
public policy issue, such as economic policy making in the case of climate
change, with large nwnbers of interest groups and different value systems,
different cognitive frameworks (for example different scientific, economic and
public policy theories), different spatio-temporal scales and many possible
policy instruments, would underestimate the unpredictability and novelty
inherent in complex systems.
An evaluation of developments in the approach to economic policy on global
climate change reveals that it falls short in approaching as a complex and dynamic
environmental problem in the first place, as would be evident through an
interactive policy learning process. Although the complexities of the problem are
more and more intemaiised, there is no real learning process that places these ideas
into the right context. The decision analytical tools that are developed are often
focused on complex, but closed systems. In addition, normative criteria, such as
economic efficiency or sustainability, can play an important role in shaping the
debate, but it is evident that the debate on economic policy for global change often
resorts to counterproductive theoretical infighting as based on different a priori
chosen criteria. The selection of criteria for further analysis, however, should be
the result of an interactive learning process, as reality is larger than these theories
can comprehend It is therefore recommended that the appHcatioD of interactive
learning frameworks to real-world economic policy making for complex and
dynamic environmental problems is an area for further research.
ENDNOTES
This paper is largely based on the author's D.Com thesis and funding
from the CSIR is gratefully acknowledged.
2 For a more detailed discussion see Kruger & de Wit (2002).
3 Hanna et al. (1997) applied the institutional focus on property right
structures to ecological problems, but this is not the general case in the
subject field of ecological economics.
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
SAJEMS NS Vol 5 (2002) No 2 331
4 See Prigogine and Stengers (1984), Stewart (1990) and Waldrop (1994)
for background infonnation.
5 For a comprehensive review see de Wit (2001 :141-53)
6 Root defmitions are concise verbal definitions expressing the nature of
purposeful activity systems regarded as relevant to exploring the problem
situation. A full root definition would take the fonn: do X by Y in order
to achieve Z.
7 For a comprehensive discussion on these issues see de Wit (2001: 155-9).
8 I am grateful to an anonymous reviewer who alerted me to this point.
REFERENCES
ANDERSON, T.L. & HILL, P.I (1995) Wildlife in the Marketplace. The
Political Economy Forum, Lanham, Maryland: Rowman & Littlefield.
2 ANDERSON, T.L. & LEAL, D.R. (1991) Free Market Environmen-
talism, Pacific Research Institute for Public Policy, San Fransisco,
Boulder etc.: Westview.
3 BLAUG, M. (1980) The Methodology of Economics, Cambridge:
Cambridge University Press.
4 BLAUG, M. (1992) The Methodology of Economics, Cambridge:
Cambridge University Press.
5 BOADWAY, R. (1974) "The Welfare Foundations of Cost-Benefit
Analysis", Economic Journal, 84: 926-39.
6 BOLAND, L.A. (1982) The Foundations of Economic Method, London
etc.: Allen & Unwin.
7 BRENNAN, G. & BUCHANAN, J.M. (1985) The Reason of Rules,
Constitutional Political Economy, Cambridge etc.: Cambridge University
Press.
8 BROMLEY, D.W. (1991) Environment and Economy. Property Rights
and Public Policy, Oxford, UK & Cambridge, USA: Basil Blackwell.
9 BROMLEY, D.W. (2000) "The Two Realms of Reason: Calculation and
Sentiment", Paper presented at the 7!l! Ulvon conference on
Environmental Economics, Ulvon, Sweden, June 18-21.
10 BRUCE, J.1., LEE, H. & HAITES, E.F. (eds.) (1996) "Climate Change
1995. Economic and Social Dimensions of Climate Change".
Contribution of Working Group III to the Second Assessment Report
(SAR) of the Intergovernmental Panel on Climate Change (IPCC).
Cambridge: Cambridge University Press.
II BUCHANAN, 1M. (1987) Economics. Between Predictive Science and
Moral Philosophy, compiled by R.D. Tollison. & V.I. VanBerg. College
Station: Texas A&M University Press.
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
332 SAJEMS NS Vol 5 (2002) No 2
12 BUCHANAN, I.M. (1991) Constitutional Economics, Cambridge, Mass.:
Basil Blackwell.
13 BUSTERUD, J. (l977) "IIASA and the Policy Maker: Integrating the
Research Function into Decision Making Research Memoranda", RM-77-
045, International Institute for Applied System Analysis (IIASA).
14 CALDWELL, B. {I 982) Beyond Positivism, London: Allen & Unwin.
15 CHECKLAND, P. (1981) Systems Thinking, Systems Practice,
Chichester: Wiley.
16 CHECKLAND, P. & SCHOLES, J. (1990) Soft Systems Methodology in
Action, Chichester: Wiley.
17 CLARK, N., PEREZ-TREJO, F. & ALLEN, P. (1995) Evolutionary
Dynamics and Sustainable Development. A Systems Approach, Aldershot,
UK & Brookfield, US: Edward Elgar.
18 CLINE, W.R (1992) The Economics of Global Warming, Washington,
D.C.: Institute for International Economics (lIED).
19 COMFORT, L.K. (1999) Book review: "Taking Complexity Seriously:
Policy Analysis, Triangulation, and Sustainable Development", Journal
of Policy Analysis and Management, 18(1): 181-4.
20 CUMBERLAND, J.H. (1994) "Ecology, Economic Incentives, and
Public Policy in the Design of a Transdisciplinary Pollution Control
Instrument", in Toward Sustainable Development. Concepts, Methods
and Policy, edited by J.C.J.M. van den Bergh & J. van der Straaten.
International Society for Ecological Economics (lSEE). Washington,
D.C.: Island Press.
21 DALY, H.E. (l992) "Allocation, Distribution, and Scale: Towards an
Economics that is Efficient, Just and Sustainable, Ecological Economics,
6: 185-93.
22 DALY, H.E. (1999) "Reply to Marcus Stewen", Ecological Economics,
30: 1-2.
23 DE GREENE, K. (1993) A Systems-Based Approach to Policy Making.
Boston: Kluwer.
24 DE WIT, M.P. (2001) "Economic Policy Making for Complex and
Dynamic Environmental Problems: a Conceptual Framework", D. Com.
thesis, University of Pretoria, Pretoria.
25 DROR, Y. (l971) Design for Policy Sciences, New York: American
Elsevier.
26 EICHNER, A.S. (ed.) (l983) Why Economics is Not Yet a Science.
London: MacMillan.
27 FANKHAUSER, S. (1995) Valuing Climate Change: The Economics of
the Greenhouse, London: Earthscan.
28 FRIEDMAN, M. (1953) Essays in Positive Economics, Chicago:
University of Chicago Press.
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
SAJEMS NS Vol 5 (2002) No 2 333
29 GARDINER, P.e. & FORD, A. (1980) "Which Policy Run is Best, and
Who Says So?" in System Dynamics: TIMS Studies in the Management
Sciences, A.A. Legasto, Forrester, J.W. & Lyneis, J.M. (eds.) Amsterdam
etc.: North-Holland.
30 GILL, R. (1996) "An Integrated Social Fabric Matrix/Systems Dynamics
Approach to Policy Analysis", Systems Dynamics Review, 12(3): l67-8l.
31 GUNNINGHAM, N., & GRABOSKY, P. (with Sinclair, D.) (1998)
Smart Regulation. Designing Environmental Policy, Oxford: Claredon.
32 HAHN, F. & HOLLIS, M. (eds.) (1979) Philosophy and Economic
Theory, Oxford: Oxford University Press.
33 HAHN, R.W. (1989) "A Primer on Environmental Policy Design",
Fundamentals of Pure and Applied Economics. Government Ownership
and Regulation of Economic Activity Section, vol. 34, Chur, etc.:
Harwood Academic Publishers.
34 HANNA, S., FOLKE, C. & MALER, K-G. (eds.) (1996) Rights to
Nature: Ecological, Economic, Cultural and Political Principles of
Institutions for the Environment, Washington, D.C.: Island Press.
35 HIRSCHMAN, A. & LINDBLOM, e. (1972) "Economic Development,
Research and Development, Policy Making: Some Convergent Views", in
Systems Thinking, Emery, F.E. (ed.). Harmondsworth: Penguin.
36 INTERNATIONAL PANEL ON CLIMATE CHANGE (IPCC), (2001)
"Climate Change 2001", Contributions of Working Group ill to the Third
Assessment Report of the IPCe. Cambridge: Cambridge University
Press.
37 IN'T VELD, R.I. & KRAAN, D.I. (1991) "Introduction", in
Environmental Protection: Public or Private Choice, (eds.) by D.I. Kraan
& R.I. in't Veld. Dordrecht: Kluwer.
38 KEEN, P.G.W. & SCOTI-MORTON, M.S. (1978) Decision Support
Systems, New York: Addison-Wesley.
39 KELLEHER, G.I. (ed.) (1970) The Challenge to Systems Analysis;
Public Policy and Social Change, New York etc.: Wiley.
40 KEYNES, J.N. (1917) The Scope and Method of Political Economy,
London: A.M. Kelley.
41 KOLB, D.A. (1984) Experiential Learning, Englewood Cliffs, N.I.:
Prentice-Hall.
42 KRUGER, A.I. & DE WIT, M.P. (2002) "A Philosophical Inquiry into
Sustainability Assessment as Applied to Agriculture", Paper presented a
Forum for Economics and Environment (FEE) conference, V &A
Waterfront, Cape Town, 11-12 February.
43 LINDBLOM, C.E. (1959) "The Science of Muddling Through", Public
Administration Review, 19(2): 79-88.
44 LINDBLOM, C.E. (1968) The Policy Making Process, New Jersey:
Prentice-Hall.
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
334 SAJEMS NS Vol 5 (2002) No 2
45 LINDBLOM, C.E. (1979) "Still Muddling, not yet Through", Public
Administration Review, 39: 517-26. .
46 MARCH, J.G. & SIMON, H.A. (1958) Organization, New York: Wiley.
47 MURTHY, P.N. (2000) "Complex Societal Problem Solving: A Possible
Set of Methodological Criteria", Systems Research and Behavioural
Science, 17(1): 73-101.
48 NORDHAUS, W.D. (1991) "To Slow or Not to Slow: the Economics of
the Greenhouse Effect", Economic Journal, 101, July: 920-37.
49 OPSCHOOR, H. & Turner, K. (eds.) (1994) Economic incentives and
environmental policies: principles and practice. Dordrecht etc.: Kluwer.
50 PEET, 1. 1992. Energy and the ecological economics of sustainability.
Washington, D.C. & Covelo, California: Island Press.
51 POLLARD, E. & LIEBECK., H. (1994) The Oxford Paperback
Dictionary. (4th ed.) Oxford etc.: Oxford University Press.
52 PRIGOGINE, I. & STENGERS, I. (1984) Order Out of Chaos. Man's
New Dialogue with Nature, New York: Bantam.
53 QUARLES VAN UFFORD, P. (1993) "Knowledge and Ignorance in the
Practices of Development Policy", in An Anthropological critique of
Development: The Growth of Ignorance, Hobart, H. (ed.) London:
Routledge.
54 RASTOGI, P.N. (1992) Policy Analysis and Problem-Solving for Social
Systems - Toward Understanding, Monitoring, and Managing Complex
Real World Problems, Sage.
55 ROE, E.M. (1998) Taking Complexity Seriously: Policy Analysis,
Triangulation, and Sustainable Development, Boston: Kluwer.
56 SAEED, K. (1994) Development Planning and Policy Design: A Systems
Dynamics Approach, Aldershot etc.: Avebury.
57 SAEED, K. (1998) Towards Sustainable Development: Essays on
Systems Analysis of National Policy, (2nd ed.) Aldershot etc.: Ashgate.
58 SCHMOLDT, D.L. & PETERSON, D.L. (1999) "Analytical Group
Decision Making in Natural Resources: Methodology and Application",
Forest Science, 46(1): 62-75.
59 SCHNEIDER, A.L. & INGRAM, H. (1997) Policy Design for
Democracy, University Press of Kansas.
60 SCHOOT-UlTERKAMP, AJ.M. (1999) "Book Review: Taking
Complexity Seriously: Policy Analysis, Triangulation, and Sustainable
Development", Journal of Environmental Policy and Planning, 1 (3): 266-
7.
61 SEN, A.K. (1970) Collective Choice and Individual Welfare, Amsterdam:
North-Holland.
62 SENGE, P. (1990) The Fifth DiSCipline: The Art and Practice of the
Learning Organisation, New York: CurrencylDoubleday.
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.
SAJEMS NS Vol 5 (2002) No 2 335
63 SIMON, HA. (1977) The New Science of Management Decision,
Revised ed. Englewood Cliffs, NJ: Prentice-Hall.
64 STACEY, R (1995) "The Science of Complexity: An Alternative
Perspective for Strategic Change Processes", Strategic Management
Journal, 16(6): 477-95.
65 STERNER, T. (1994) "Policy Instruments for a Sustainable Economy",
in Economic Policies for Sustainable Development, Sterner, T. (ed.).
Dordrecht: Kluwer.
66 STEWART, I. (1990) Does God Play Dice? The New Mathematics of
Chaos,London:Pengum.
67 THOMAS, J.W. & GRINDLE, M. (1990) "After the Decision:
Implementing Policy Refonns m Developmg Countries", World
Development, 18(8), August: 1163-81.
68 UNITED NATIONS DEVELOPMENT PROGRAMME (UNDP) (1999)
Policy Analysis and Formulation for Sustamable Livelihoods, Internet:
http://www.undp.org/sVdocurnents/strategy-paper .. .Ipolicy _analysis_and
jonnulation.html, Access: 2 June 2000.
69 VENNIX, J.A.M. (1999) "Group Model-Buildmg: Tackling Messy
Problems", Systems Dynamics Review, 15(4): 379-401.
70 WALDROP, M.M. (1994) Complexity. The Emerging Science at the
Edge of Chaos and Order, London: Pengum.
71 WARD, B. (1972) What's Wrong with Economics? New York: Basic
Books.
72 WARFIELD, J.N. (1994) A Science of Generic Design: Managing
Complexity through Systems DeSign, (2nd ed.) Iowa State University
Press! Ames.
73 WARFIELD, J.N. (1999) "Twenty Laws of Complexity: Science
Applicable m Organisations", Systems Research and Behavioural
Science, 16(1); 3-40.
74 WOLFENDEN, J. Policy Analysis for ecolOgically Sustainable
Development Research outline for Ph.D. Internet: http://www-
personal.une.edu.aul-jwolfendljawsphd.htIn. Access: 23 August 2000.
R
ep
ro
du
ce
d
by
S
ab
in
et
G
at
ew
ay
u
nd
er
li
ce
nc
e
gr
an
te
d
by
th
e
P
ub
lis
he
r
(d
at
ed
2
00
9)
.