Chapter 13

Environmental Policy Instruments

Gjalt Huppes and Udo E. Simonis

Contents 13.1 Introduction

13.1 Introduction ............................................................ 239
13.1.1 Approaches to Governance ...................................... 239 13.1.1 Approaches to Governance
13.1.2 Outline of the Chapter.............................................. 241
13.2 Policy Instruments in
a Long-Term Perspective ....................................... 244 What is the role of environmental policy instruments?
13.2.1 The Future Context of Instruments .......................... 244 In simplified terms, environmental policy instruments
13.2.2 General Tendencies in Society ................................. 244 can be said to link policy development and decision-
13.2.3 Globalisation Tendencies ......................................... 247 making to policy implementation. Starting from policy
13.2.4 General Tendencies in
Environmental Problems .......................................... 248 development, the policy problem is translated into
13.2.5 Prospects for Instruments......................................... 250 operational goals, the appropriate instruments are cho-
13.2.6 Strategic Instrument Choices Ahead ........................ 251 sen, and their implementation achieves the goals. This
13.3 Policy Instruments: What picture of policy as a linear, stepwise activity is an over-
Are They Good for? ............................................... 254 simplification. For instance, the definition of the policy
13.3.1 Policy Instruments: What Are They?............................ 254 problem is often already based on instrument choice,
13.3.2 Why Have Policy Instruments?................................ 255
13.3.3 A Framework for Analysing
resulting in a circularity that disrupts the seemingly
Policy Instruments ................................................... 255 rational linear picture. Permits for the operation of
13.3.4 Policy Instruments in Context .................................. 257 installations tend to define problems in terms of the
13.3.5 Environmental Problems: effects the installation has on its environment, for
Causes and Solutions ............................................... 259
13.3.6 Evaluation Criteria for Policy Instruments .............. 261
instance in terms of noise, stench and eutrophication of
a nearby lake. Having defined the problem in this way,
13.4 Design, Analysis and Evaluation
the choice of policy instrumentation is then more or
of Policy Instruments ............................................. 264
13.4.1 A Framework for Design and Analysis.................... 264 less limited to variants of the permit, as emission taxes
13.4.2 A Thousand Instruments Defined, and liability rules simply do not fit the problem defini-
with Examples.......................................................... 271 tion. Such more abstract instruments require a more
13.4.3 Choice of Instruments in Policy Design .................. 273
aggregate view of the problem, involving for instance
References ........................................................................... 279 groups of installations or activities (see Huppes 1993).
It is also possible to take a more complex view of
the policy process, starting with the social processes
involved. Policy implementation presupposes a par-
ticular context in which the instruments function,
involving public and private organisations and many
individuals, each with their own roles, knowledge and
ideas. In some countries, installation permits are speci-
G. Huppes () fied and become legally binding without having much
Institute of Environmental Sciences at Leiden University
(CML), The Netherlands effect on behaviour. Actual behaviour may deviate
e-mail: huppes@[Link] from the permit regulations because monitoring is

Jan J. Boersema and Lucas Reijnders (eds.), Principles of Environmental Sciences, 239
© Springer Science + Business Media B.V. 2009
240 G. Huppes and U.E. Simonis

lacking, officials agree to allow deviations, or sanc- eutrophication, seems to create an impulse towards
tions cannot be effectively applied. In such quasi- effective policy implementation. The two views also
illegal, seemingly ineffective regulatory situations, meet at a more strategic level. In formalised bureau-
however, private organisations may still decide to cratic societies, like the United States, the rigidity of
introduce environmental improvements, on a variety of regulations and their increasing complexity exerts a
grounds, and may communicate and coordinate their pressure towards more informal procedures.
actions with officials. These may then sanction the Conversely, the lack of effective regulation in more
actual behaviour, either informally or by ultimately informal societies, as England and the Netherlands
issuing a formal permit befitting this behaviour. Such used to be, has led to pressures towards more formal
an informal style of regulation has been dominant in procedures, as have the formal rules introduced by
England and the Netherlands, where officials commu- the EU. The integration of process aspects and instru-
nicate with firms and agree on actual operations mentation requires a more abstract approach to the
without necessarily issuing a permit. As a Dutch envi- policy development and implementation process.
ronment officer once expressed it: ‘As long as there is German authors (especially Mayntz 1978; Luhmann
no permit, I can influence my firms. As soon as the 1989) have been highly influential in this respect,
permit is issued officially, my influence is gone.’ In also in the US, linking the Weberian bureaucracy
such situations, there is no clear distinction between theory with political process theory into modern
the setting of policy goals, instrument choice and pol- administrative sciences.
icy implementation, let alone that they occur in this How does this chapter deal with such fundamental
precise sequence. A context in which such officials differences of view when discussing the subject of
have all sorts of, preferably broad and unchecked, policy instruments? We have decided to apply both.
powers allows them to negotiate. It is these operational For some instrument types, this chapter uses the hier-
practices in society, embedded in culture and institu- archical vertical model, mitigated by the knowledge
tions, which may largely determine policy develop- that policy development and implementation is a
ment and policy instrumentation. In a more legalistic cyclic process and that context, in terms of culture
culture, as in the US, such informal procedures may and institutions, is a decisive factor in the functioning
play a limited role. of such instruments, especially in situations where
These two views of the environmental policy pro- horizontal government is the dominant mode. At the
cess relate to two basic views on the way society can other end of the scale, there are policy-making pro-
be influenced, one vertical and the other horizontal. cesses in which consultations between those involved
The vertical view links a formal democracy which create a consensus between them which hardly needs
sets policy goals with a Weberian ideal type of hierar- formalising. A covenant, without any sanctions for
chical bureaucracy implementing these goals (see non-compliance, may be the only externally visible
Weber 1947). The horizontal view is associated with instrument in this situation. In between, there is a
direct democracy and social integration through range of instruments which guide private choices
meso-level bodies, like trade associations. Covenants without ever specifying the behaviour of the actors
between all public and private parties involved set involved in any detail. Emission taxes and environ-
environmental goals and leave their implementation mental information are examples of situations in
to those most directly concerned. Even if one does which very limited public actions have a far-reaching
not specifically opt for one of these views, one still influence on many decisions. Taxes on petrol in
has to deal with these diverging aspects, as ignoring Europe and Japan have led to much lower gasoline
social aspects may well make enacted policy ineffec- consumption and related emissions than in the US,
tive. In the Netherlands, three decades of formal reg- while subsidies on energy in the former communist
ulations on the manure problem have been to no avail, countries, still not fully being broken down after the
as local governments refused to effectively imple- collapse of communism, have led to extremely high
ment the successive laws enacted (see Huppes and energy consumption per Euro of national income.
Kagan 1989). Pressure by the EU in this field, with Such mechanisms work regardless of the intentions
heavy penalties being imposed on governments for with which the subsidies and taxes were created; it is
not meeting quality standards for air quality and through anonymous markets that they exert their
13 Environmental Policy Instruments 241

influence. The middle European countries now hav-

ing become members of the EU have adapted to world Policy Policy Policy
development instrumentation implementation
market prices and Western European taxing systems
and are transforming rapidly to the Western European
energy consumption levels per Euro of national
income. Fig. 13.1 The place of policy instruments in the policy
process
Furthermore, this chapter covers not only the ver-
tical and horizontal approaches to public policy, but
also purely private developments. An example is that It is now widely accepted that human society has to
of the international standards for environmental per- find a sustainable way of dealing with its environment.
formance specified in the ISO 14000 Series. ISO The basic ethical position taken in sustainability is one
(International Organisation for Standardisation) is a of distributive justice within and between generations.
private organisation. These standards, in combination The implications for empirical analysis and social
with cultural values and liability rules, are having a action are enormous, and in terms of concrete actions
profound influence on corporate behaviour in particu- obviously depend on the more detailed ways in which
lar. The inclusion of such non-public policy instru- the sustainability concept may be specified. Somehow,
ments in the analysis results in a broad view on society has to incorporate and maintain feedback
instruments. mechanisms which redirect the actions of many, a pro-
Finally, we had to decide which environmental cess which Luhmann (1989) calls autopoiesis. The
subjects to cover. In principle, this chapter includes all development of policy instruments is part of the insti-
subjects for which policy formation has developed in tutional development of society towards the kind of
such a way that clearly specifiable domains of social self-steering which is required for long-term human
activities are linked to states of the environment with welfare for all, and ultimately even for human survival.
real or potential undesirable effects, either directly on See Fig. 13.1 for a main conceptual framework. The
the environment or through the environment on soci- institutional development required, often referred to as
ety. Hence, all environmental problems relating to ecological modernisation, centres around the develop-
emissions and a-biotic extractions are extensively ment of new policy instruments, going from curative to
covered. To some extent, the chapter also covers biotic preventive, from centralised to decentralised, and from
extractions, such as fishery rights in common waters over-regulated to stimulating (see van Tatenhove and
and quality control on sustainable forestry. Other areas Leroy 2000).
are not extensively covered in this chapter. Policy
development for land-use related subjects has not yet
developed much beyond zoning laws. Globalisation
has fundamental ecological and evolutionary conse- 13.1.2 Outline of the Chapter
quences, for instance through the spreading of species
to all ecosystems in our globalising world, from The chapter moves from the general to the specific, in
viruses and bacteria to plants, fungi and animals. Its that it starts with broad societal developments at a
effects on biodiversity will be substantial, while health global level, then goes into the social contexts of
risks may result from recombination of viruses and instrument functioning, and finally discusses the
bacteria. However, no preventive instruments are detailed specification of instruments, analysing their
available other than border controls, which are steadily strengths and weakness. Section 13.2, entitled ‘Policy
diminishing. Similarly, amenity values of landscapes, instruments in a long-term perspective’, investigates
including aspects of cultural history, cannot yet be the global social context from a long-term strategic
broadly and systematically covered by instruments, as perspective. As choices on policy instrumentation
building regulations can only partly affect such bind society for a long period, long-term changes in
amenity aspects. Clearly, then, there are domains of context have to be taken into account, in terms of
environmental policy in which steering has not yet structural, cultural, economic and political develop-
become sufficiently routine and formalised to allow ments. Major structural developments include global-
institutionalised instruments to be used. isation, which changes all organisations in business,
242 G. Huppes and U.E. Simonis

science and technology. Small countries have little Finally, Section 13.4, entitled ‘Design, analysis and
decisive influence on technologies which have been evaluation of policy instruments’, specifies the main
developed elsewhere. A major cultural trend is towards dimensions for defining instruments. It shows that it is
individualisation, with intermediate organisations like not at all clear how policy instruments can be classified
churches, trade unions and trade organisations loosen- and described. Nor is it clear how a consistent evalua-
ing their traditional grip on individuals. A key eco- tion of policy instruments can be set up. Still, as some
nomic characteristic is the continuing growth of ordering is necessary for instrument development and
income, based on structural developments at a global instrument choice, an analytic framework is developed,
level like better education, technological progress covering not only the traditional regulator-regulatee
through ongoing investments in science and technol- relations but also instruments structuring the relations
ogy and increased labour participation not yet offset between various governmental organisations, and
by shorter working hours. Such fundamental develop- instruments structuring relations between private
ments are first surveyed, with a view to finding out actors, both individuals and organisations. Some prac-
how instrument choices can take such major contex- tical guidelines are given for policy development at a
tual developments into account. A number of strategic case level.
criteria for instrument design and evaluation emerge Before discussing first strategies, then the contex-
from this analysis, which takes for granted certain tual framework and next the actual instrument analy-
rough ideas about what policy instruments are and sis, let us first try to visualise this complex subject as
why we need them. a whole, to gain some perspective (see Fig. 13.2). The
Section 13.3, entitled ‘Policy instruments: what are starting point is some sort of policy goal that is to be
they good for’, discusses basic analytics, in a bottom- achieved. In designing instruments, one first has to
up approach. Why do we need instruments for envi- decide on the actors involved, like regulators and
ronmental policy? Could integrated policies, without regulatees, but also the broader groups involved,
specific instruments, not cover all problems? As instru- which together form the stakeholders to the policy.
ments are placed in their administrative setting, while Next, the goal of the policy has to be defined at an
emphasising ‘horizontal governance’ – in which pub- operational level, like an emission volume, an immis-
lic and private partners together design solutions to sion concentration, an immission load or an allow-
common problems – the question why we need them able health risk. The next element required is something
crops up again. The answer given here is that they are that can be regulated, something that is the object of
needed for environmental effectiveness, for the simpli- regulation, like an installation, a product or a behav-
fication of the policy process, and for building into the iour. Finally, something is needed that sets the policy
fabric of society the safeguards for long-term sustain- in motion, such as an operational influencing mecha-
ability. After these hurdles have been taken, the analy- nism. Some positive or negative incentive has to be
sis is constructed around the theme of the nature of created, like a prohibition, a subsidy or the obligation
environmental problems and the general mechanisms to display information, to act as the instrument’s
that cause undesirable environmental consequences. influencing mechanism. What happens next, empiri-
Concepts like ‘external effects’, ‘collective goods’ and cally speaking, is set in motion, that is, caused, by the
‘free rider problems’ are surveyed, as these constitute instrument.
the particular context in which instruments should pro- This then leads on to an empirical part of the analy-
vide solutions. A final introductory theme is that of sis, involving all sorts of mechanisms in society which
evaluating alternative instruments for environmental may have direct effects, involving simple causal chains
policy. A distinction is made here between first-order in the short term; indirect effects, taking into account
criteria like effectiveness and costs; second-order cri- longer effect chains; or secondary effects, involving
teria like requirements on administrative capacity and broader effect mechanisms. The direct effect of an
effects on technology development, which are mostly energy tax, for instance, is a change in energy prices and
difficult to quantify; and strategic third-order criteria consumption, while its indirect effect is increased spend-
relating to instruments, such as fitting them into over- ing on non-energy expenditures and a shift towards
all regulatory and broader institutional developments, more energy-efficient products, and its secondary effects
which are never quantified. may involve adjustments to the goals of research and
13 Environmental Policy Instruments 243

Fig. 13.2 Design, empirical analysis and STRATEGIC DESIGN

evaluation of policy instruments CONSIDERATIONS

1 2 cultural; institutional;
ACTORS OPERATIONAL
economic; political
INVOLVED GOALS
DESIGN
of 3 4
instruments REGULATED INFLUENCING
and policies OBJECTS MECHANISMS

FURTHER EFFECT MECHANISMS

IN SOCIETY
• direct
• indirect
• secondary
EMPIRICAL
ANALYSIS
of EFFECT MECHANISMS
instruments IN/T HROUGH ENVIRONMENT:
and policies • direct
• indirect
• secondary

CHECK on:
• tragedy of the commons
• prisoner’s dilemma
• free rider problem
• problem shifting
NORMATIVE
EVALUATION
of EVALUATION, using:
instruments • first-order criteria
and policies • second-order criteria
• strategic criteria

Legend:

effect information line of

mechanisms for evaluation reasoning

development in large sections of society. This empirical accepted principle that those who cause damage have
analysis of instrument application in society is then fol- to pay for it? The highest level of strategic consider-
lowed by an empirical analysis of the resulting environ- ations takes into account the expected developments in
mental effects, with feedback to society. our world, relating to trends like globalisation. In a
A third element in the analysis is evaluative: is it world where international trade becomes more perva-
worthwhile to use the instrument, reckoning with its sive, national regulations increasingly lead to loss of
implementation costs and effects? This question is competitiveness. Instruments differ in this respect, with
related on the one hand to the empirical functioning emission taxes having major effects and consumption
of the instrument, involving evaluation criteria like changes minor ones, and instruments may be designed
environmental effectiveness, cost and regulatory effort. in such a way as to prevent this unintended conse-
On the other hand, an important set of evaluation crite- quence. The empirical and normative analyses feed
ria is of a more abstract nature. How do subsidies for back to instrument choice and policy design. The empir-
not polluting relate to liability law, where it is an ical analysis is of course not based on actual empirical
244 G. Huppes and U.E. Simonis

performance, which is not yet there, but on expected strategic and tactical aspects of technology develop-
consequences. This completes the policy instrumenta- ment has been reduced to virtually zero in most per-
tion cycle (see the arrow on the left in Fig. 13.2). mitting situations. Only additional instruments like
large subsidies can exert some influence in exceptional
cases. For instance, highly efficient and environmen-
tally benign coal gasification plants for power genera-
13.2 Policy Instruments in a Long-Term tion are only built with huge subsidies now, and no
Perspective more than a few dozens of such installations have been
built in the world so far. Therefore, developing the
13.2.1 The Future Context instruments for environmental policy, as an operational
set, requires such long-term developments in society to
of Instruments
be taken into account, not only in the local firms but
also in the broader societal settings.
Policy instruments tend to bind society for a long A number of these developments are addressed
period of time, as changing them often requires com- below, with some indications of their potential signifi-
plex administrative procedures and a decision to adopt cance for the instrumentation of environmental policy.
a particular instrument in one context influences instru- Relevant and significant developments are taking place
ment application in other contexts. Given this long- in the overall structure of society, in general cultural
term character of instrument choices, the context of developments, in developments in the economy and
their functioning is not the situation in which they are specifically in industrial relations, in the changing role
being introduced but more that of the future. Although of government and, last but not least, in the changing
the future is hard to predict and also shaped by deliber- nature of environmental problems. We will also discuss
ate choices, some general developments may be speci- in some more detail the consequences of globalisation.
fied, such as structural trends. Such structural changes, The next aspect to be addressed is that of the prospects
both in society and the environment, lead to shifting for various policy instruments in a changing world.
conditions, resulting in a different functioning of exist- These prospects do not automatically lead to clear indi-
ing instruments and emerging options for new instru- cations as to which instrument is to be used, when and
ments. At the core of an instrument, there may be how. In this respect, policy instrumentation as a societal
simple social mechanisms which were rightly taken process is itself also partly developing through causal
for granted when the instrument was introduced, but mechanisms of a social nature, and is partly based on
are not self-evident in the long run. For instance, the explicit strategic considerations and decisions. Some
workhorse of environmental policy in England and the major strategic choices in policy instrumentation are
Netherlands is the permit for individual installations. It identified in the final section of the chapter, which
was assumed that decisions on how to operate these approaches the arena of political discussion.
could be discussed with those responsible before the
installations were designed and built, and long before
a permit was due. The firm’s planning was geared to
that of the permitting procedure, with quite some pro- 13.2.2 General Tendencies in Society
cedural flexibility. In many sectors, the periods for
planning and implementation technologies have now The social structure of Western countries is showing a
been reduced to months rather than years, and involve trend towards a decreasing role of intermediate organi-
integrated decisions on technologies implemented sations like churches, trade unions, parties and clubs.
worldwide. Discussions with the local environmental Institutional integration is weakened and its indirect
authority will only relate to some of the details, but not control over individuals is reduced. This structural
to the technology itself. Nor are consultations allowed development reflects a double cultural development.
to address factors involving real cost shifts, as expected As early as the 1950s, Riesman (1950) described a
costs have already been incorporated in the decision- change from inner directed to other directed control of
making process in the network of firms involved in the people’s behaviour. This means that it is not interna-
technology. Thus, the role of local authorities in the lised norms and values that guide specific choices, but
13 Environmental Policy Instruments 245

the notions of others about this subject as determining knowledge embodied in any given product is expand-
his choices. Moral precepts are replaced by self-cen- ing and the technological complexity in or behind most
tred considerations of expediency, based on how oth- products is increasing. The innovative capacity of firms
ers view an action and react to it. General considerations is increasingly based on functionally differentiated,
on what is right and wrong lose their place in guiding more or less independent innovation generating organ-
actions. The second tendency, stated even longer ago isations. At the same time, the organisation of success-
by Tönnies (1887), relates to the question who these ful firms is more open to external innovation options.
others may be. The reference group for one’s norms A large firm like Shell has incorporated its main
and values is shifting from a closely knit small research capacity in an independent organisation,
group of partners for life, one’s ‘Gemeinschaft’, to called Global Solutions, which has Shell as its main
larger groups of more shifting acquaintances, the client, but operates on the world market. The market
‘Gesellschaft’. These basic developments provide for innovations in industrial production has increas-
opportunities for new forms of integration. The differ- ingly gone global, including and relying on the infor-
ences between national cultures are diminishing, as a mation and service industries. It is here that future
global culture is developing through shared TV pro- technologies with their specific environmental conse-
grammes, advertisements, books and movies, through quences are born and start to diffuse throughout the
the nearly universal marketing of products and through world. Technological innovation, viewed by
internationalisation of contacts via tourism and the Schumpeter (1942/1976) half a century ago as the
Internet. Although some debate is possible on new capacity of owners of firms to innovate their own activ-
modes of small group integration, the tendencies spec- ities, has now become a more or less independent
ified above all lead in the same direction, in which nor- capacity at the service of all other firms. It is an open
mative control on individuals in their roles as consumers question whether ‘Schumpeterian dynamics’ and the
and producers is decreasing. The legitimacy of mea- Rio imperative of sustainability can be made compati-
sures founded in morality has been reduced in the pro- ble, based on the development of zero emission or
cess, and no longer plays its invisible (quasi-automatic) Factor Ten technologies which will not arise spontane-
role the way it used to. Therefore, the assumption that ously. If firms want to invest in new installations, they
rules will be followed more or less automatically may will often be faced by a stark choice: here and now, or
increasingly come to be questioned. Although cultural not here at all. Delays due to environmental permitting
controls on behaviour are shifting to higher-level col- procedures are becoming increasingly unacceptable to
lectivities, the subjective view at a global level is ulti- a firm which may just have acquired new and superior
mately that of increasing individualism, with people technology from the specialised technology develop-
deciding on ever more subjects for themselves, arguing ers abroad, with only a few weeks or months of head
from their individual views of the world. This theme of start on its competitors. In this new situation, traditional
individualisation has been extensively discussed by regulatory controls on technologies clearly have to be
Beck and is summarised in Beck and Beck-Gernsheim redefined, as they can no longer be based on insights by
(2002). In a world dominated by relatively superficial the regulators into the firms they are regulating.
types of information, the reference for individual deci- Another structural change in the economy is the
sion-making may be quite poor and simple, relating to shift from product to service. This development also
individual experiences with simple generalisations. affects traditional hardware, such as cars that are not
This de-socialising tendency (with integration levels bought but increasingly being leased. In addition, it
becoming less binding and content-poor, compare is embodied in the emerging information and com-
Durkheims anomie) may be countered by other struc- munication technologies, whose hardware is mainly
tural mechanisms, like the globalisation of science and owned by providers, while clients pay for services
literature. It is difficult, however, to link these global only. This development is taken one step further
activities to new types of normative and broader cul- when firms do not provide services but act as ser-
tural integration. vice-providing organisations, leaving the actual
The global economy is going through an extensive physical activities related to service provision to
structural change, based on new technologies and new smaller units. Franchising is one example that has
forms of communication. The amount of specialised been in existence for some time, but in industrial
246 G. Huppes and U.E. Simonis

production, designers and marketers now externa- environment? Long-term developments in this respect
lise production to a much higher degree than they are pointing in the opposite direction. Economic
used to. A fast growing firm like Nokia is an exam- growth is fairly constant in the long run, with real
ple, in which suppliers of all major parts of mobile growth in the last decade being of the order of 3% per
phones and physical networks are being chosen flex- year, doubling consumption every 23 years, while
ibly every few months. The actual service provision international trade is increasing by 7% per year. The
is by competitive network operators, which ensure implied growth in labour productivity is not matched
that the phones function for the consumers who are by a proportional decrease in total labour time. On the
paying them. And if services are not externalised contrary, there are pressures towards longer working
outside the firm, larger organisations set up business hours per worker. This means that total production will
units, which in many respects function as indepen- rise, as will the physical activities related to negative
dent firms. They sell and deliver their goods and ser- environmental effects. The shifting nature and compo-
vices to other business units in the organisation but sition of consumption do not indicate an automatic
also to the outside markets, while other business solution to environmental problems. For instance, ser-
units are not obliged to buy their inputs from busi- vices may seem attractive in that they are free of mat-
ness units within the same firm. No doubt, this ter. However, the system activities required for growing
downsizing of organisations, combined with services like mobile phoning and international holi-
increased international competition for more or less days are highly energy-intensive and require substan-
standardised activities, makes environmental con- tial material inputs of a very specialised nature.
trol on such standardised mass production activities Tendencies in government reflect these broader
very hard for individual countries. The non-stan- developments in structure, culture and economy. The
dardised, creative and strategic aspects of the inno- ideal of planning the future is dead. Although it is still
vation process may hardly have discernable reflected in names like the Central Planning Bureau in
environmental effects, and are not influenced by the Netherlands, planning, even the sort of indicative
binding instruments or market instruments, leaving planning that existed in France, has disappeared both
the scene to ‘softer’ informational instruments and from government and from business. (Mintzberg, the
to structural instruments with a limited environmen- ‘guru’ of strategic planning in the firm, has named one
tal scope. of his books ‘The end of strategic planning’, Mintzberg
A shift in a similar direction is the changing nature 1994). Of course, targets are still being set, such as the
of coordination in the economy. There is a marked reduction targets for greenhouse gas emissions in the
shift from hierarchical control to coordination by con- Kyoto protocol, and the targets in national policy plans.
tracts and markets, quite in line with the tendency However, these targets are not part of a broad vision on
towards individualisation. Contracts and markets are societal development, and the link between the quanti-
not fully anonymous but involve flexible relations, tative goal and implementation activities is decidedly
increasingly based on global communication networks. weak. In fact, global implementation mechanisms in
Anything can be bought anywhere in the world at short the form of policy instruments are largely lacking.
notice. The main fixed points are the locations of con- Although some of the vocabulary is thus still there,
sumers and most employees and some bulk resources, integrated strategic planning and control are fading. In
while most of the other aspects of the physical eco- particular, the strong control that used to be provided
nomic activities are variable in terms of location, due by the informal modes of regulation in England and
also to decreasing transport costs. Formerly the domain the Netherlands has first lost its glamour and then its
of large multinationals, now small and medium sized effectiveness. Ultimately, the informal flexibility used
enterprises are also turning to international trade, inte- to be backed up by the power of officials to implement
grating in international networks. As a consequence of what they liked, even if it was unreasonable. Negotiating
these developments, traditional national regulators ‘in the shadow of the law’ is increasingly difficult and
must feel their powers vanishing. the nature of the instruments involved has changed. It
Will there be an end to the increase in global pro- seems such countries are all shifting towards the
duction and consumption, easing the pressure on the American, more formalised and litigative procedures,
13 Environmental Policy Instruments 247

which is not very helpful in the new situation of global developments pose severe problems for national envi-
competition for technologies and products. ronmental policy formation and instrumentation.
New developments also include avoiding the com- The consequence in terms of instruments is that
plexities and costs of formalised regulation by concen- binding instruments based on command and control
trating on consensus building. Consensual processes at a technological level are becoming harder to apply.
are major vehicles for change, often with horizontal Any instrument with real effectiveness leads to costs.
government as a guiding principle. Building on avail- The idea that economic-environmental win-win situ-
able integration in society, the domain of application ations will emerge spontaneously is attractive but
of this approach will tend to become more local. In highly improbable, as the structural causes of envi-
local affairs, those involved know each other already, ronmental problems remain and environmental pres-
and horizontal government is not so much of a change. sures increase because of population growth and
For higher level problems, as environmental problems economic growth. It seems that win-win situations
increasingly are, this is a change relative to existing are related to weak sustainability, in which innova-
practice. It is a tendency that builds on corporate gov- tions are attractive, environmentally and economi-
ernment ideas of involving all major parties involved cally, per unit of product. Since, at the same time,
in a negotiating procedure, in which win-win situa- economic growth is implied, the overall effects at the
tions are created, to the advantage of all. In corporate macro level will usually be detrimental to environ-
government, it was the top echelons of the socio-eco- mental quality. We therefore assume that, for the
nomic institutions which made the deals; now it is foreseeable future, environmental policy will not be
‘those involved’ in general, that is, the stakeholders, expendable at all. Quite the opposite, in fact. Its
who together decide on some problem or action. On nature, however, will have to change.
the one hand, this tendency reflects the decreasing Therefore, effective national policies in the global
power of national governments. On the other hand, context require international co-ordination. This co-
horizontal government is also an impetus for less active ordination should not only be effected at the level of
and less binding types of regulation. The consequence setting aims, as is now increasingly the case, but to a
for environmental policy could be that for most prob- certain extent, the design of instrumentation also has to
lems, the old habit of setting standards more stringently be agreed upon internationally, if nationally effective
than those involved think reasonable is no longer an policies are to be achieved. Some early steps have been
option; only information, stimulation and financial taken in climate policy and biodiversity policy, where
incentives may remain available if this tendency international instruments between countries have been
continues. worked out for joint implementation, in the form of a
political-administrative instrument for emission trad-
ing and a clean development mechanism. For instance,
countries may now trade their emission reduction obli-
13.2.3 Globalisation Tendencies gations on a bilateral basis to improve overall effi-
ciency in emission reduction. European countries are
One recurring element in the above survey of tenden- investing in new technologies in the former communist
cies was globalisation. Environmental problems are countries to reduce emissions there relative to the
increasingly becoming transboundary or global; eco- assumed autonomous growth of emissions in these
nomic production processes integrate at a global level countries. However, any simple trading system based
for a global consumer market; and a global culture is on future emissions of a substantially fictitious nature
emerging, at least in consumption. International politi- may easily erode collective efforts and cannot easily
cal integration in blocks is losing momentum in favour be brought into line with the systematic development
of, as yet limited, integration at the global level. of regulatory instruments. Imagine that a country can
Examples of the latter are the strength of the World claim credit for helping to close down an old power
Trade Organisation (WTO) and the vehemence of the plant or helping to plant a forest in another country.
political discussion on its further expansion vis-à-vis This is now possible under joint implementation. The
social and environmental interests. Together, these net influence of these supporting activities may become
248 G. Huppes and U.E. Simonis

negligible if these developments would have taken may be discerned, each of which is important in its
place anyway. There is then no actual result, only a own right. Together, they are creating a challenge for
clear result on paper. Political deals with some finan- policy instrumentation which cannot yet be met.
cial gains to the main parties involved, may then Depletion problems of materials supply, never
become more important than real efforts for emission really high on the political agenda, would lead to a
reduction. greater emphasis on all forms of recycling. It seems
Tradable emission permits might also be imple- that the real environmental drive behind this empha-
mented as a regulatory instrument. The question is sis is not so much depletion but avoiding the detri-
then who would receive the emission rights being mental effects of primary production and final waste
traded and what the initial distribution between coun- disposal.
tries would be. Whatever the initial distribution, after Traditional environmental problems of final wastes
some time real emitters would have to have the emis- and emissions tend to become more global in nature.
sion rights corresponding to their emission volumes. The remaining ‘old’ toxicity problems are of a more
While such a system would be more transparent, it global nature, being related to substances that are not
would have serious implications in the normative set- broken down and circulate in the environment for a
up of instrumentation. Achieving global efficiency in long time, like heavy metals and pseudo-estrogens.
climate policy would mean that the trade in private Acidification and eutrophication now also tend to
emission rights would have to be preferred, on the involve larger areas, like subcontinents and coastal
same efficiency grounds now used to defend joint seas. Some of the newer problems, like ozone deple-
implementation. Broader ethical considerations tion and climate change, are essentially of a global
embodied in the normative-legal structure of most nature.
countries would favour instruments that are in line Appropriation of nature (Fischer-Kowalski et al.
with the polluter-pays principle (see OECD 1995). In 1997; Ingold 1986; Imhoff et al. 2004, going back to
this case, emission taxes are to be preferred. In the case Marx’s Das Kapital in a slightly different meaning), in
of carbon dioxide emissions in particular, such taxes terms of the share of biomass production controlled,
could in principle be set up with relative ease. The harvested and consumed by humans, is growing. Room
main choices in working out such taxes relate to who for natural type ecosystems is vanishing fast and the
receives the proceeds. If it is the national governments remaining ecosystems are controlled for their human
implementing them, this would create an incentive for oriented biomass production. Tendencies towards bio-
effective implementation. However, the carbon mass production for energy applications will give fur-
resource owning countries would effectively pay a ther impetus to this development. This will decrease
large part of the bill as a result of reduced prices for the actual volume of non-edible algae and plants and
gas, oil and coal. Ideas on legal frameworks as being of all creatures living off them – animals, fungi and
discussed in the WTO tend towards giving these coun- one-celled organisms.
tries a right to compensation. Alternatively, the emerg- The share of non-man made ecosystems is decreas-
ing global political community, potentially in the form ing, with full human control emerging except for pro-
of a World Environment and Development Organisation tected nature reserves. This leaves ever smaller patches
(see Simonis 1998) needs financing, and the carbon of ecosystems, which in principle cannot be stable in
tax proceeds would seem an ideologically acceptable terms of the number of species they can support, and
source. hence will become more dynamic.
The combined effects of the appropriation of nature
and the increasing spatial scale of human control
reduce the ecological basis of humanity not only in
13.2.4 General Tendencies
terms of volume but also in terms of quality, as many
in Environmental Problems ecosystems require large spaces for their long-term
existence.
The structural economic developments described The homogenising of ecosystems, closely related to
above have changed the nature of our relations with the the globalisation of production and the emergence of
environment (see also Box 13.1). Five basic trends global recreation, may well become a major new focus
13 Environmental Policy Instruments 249

Box 13.1 Policy instruments in a long-term perspective

Culture Economy
• Reduced role of internalised norms and values • Increasing technological complexity
• Reduced role of local groups • Functional differentiation of innovation
• Increasing role of international mass media • Shift from product to service
• Some cultural elements globalised • Shift from service provider to service organiser
Social stucture • Co-ordination: shift from hierarchies to markets
• Fewer intermediate organisations and condtracts
• Expanded liability • More flexible co-ordination in networks
• More private and national ownership of biotic Environment
resources
• Decreasing ecological resources
Government • Local problems dealt with reasonably
• Formalised and litigative tendencies in binding • Emphasis shifting to continental and global
instruments problems
• Decreasing role of binding instruments • A-biotic depletion only in very far future
• Increasing role of horizontal mechanisms, • Increasing appropriation of nature and biotic
especially locally depletion
• Uncertain role of purely private mechanisms • Continuing climate change
Population • Continuing loss of biodiversity
• Continuing growth over the next decades • Natural areas diminishing fast
• High growth rates in many poor countries • Threats of ecosystem instability
• No further growth in industrialised countries
Strategic instrument choices ahead
• Higher average age in industrialised countries
• Tradable right to environmental damage versus
Prospects for instruments collective right to unspoiled environmental
• Dynamic effectiveness of instruments: halving quality
environmental effects per unit of income • Equal right to environmental use space versus
required every 25 years equal efforts for damage reduction
• Internationally co-ordinated instrument choices • Global equity versus global efficiency
essential • Means-directed technology-specific
• Instruments for safeguarding ecological instruments versus goal-directed environmental
resources to be developed incentives
• Decreasing overall effectiveness of instruments • Normative integration of policies with broad
• Financial instruments essential for high level of internalisation in society versus political-
trade-off between economy and environment administrative debate per item of choice

for policy. As species from different ecosystems mix, pressures towards new species. Since new species of
the diversity of ecosystems is reduced. As new species higher organisms evolve over a longer time scale than
are introduced, from viruses and bacteria to fungi, plants smaller and simpler organisms, it is especially new vari-
and animals, it will be the generalists that have the great- ants of viruses and bacteria and other mono-cellular
est chance of survival, out-competing the niche special- organisms that will arise in the short term, posing threats
ists. In addition, competition between niche specialists to humans and crops, as well as to other species.
will occur, reducing the numbers of their species. As a Overall, the pressures of emissions, combined with
result, new ecosystems will evolve, with new selective the three other types of structurally increasing influence
250 G. Huppes and U.E. Simonis

on the biotic environment will lead to rapidly decreas- Such analysis instruments may have a broad influence
ing biodiversity, instability in the genomes of smaller especially if the information in the instruments is
creatures and an influence on the life support functions complemented with normative information about the
of natural ecosystems that is more difficult to specify. relative importance of different types of effect.
Policy options may focus on prevention of these mech- However, they create a limited incentive only, due to
anisms, or on the control of undesirable consequences. the collective nature of most of the environmental
Examples of the first are bans on the transport of spe- effects involved. Where real choices are to be made,
cies, which are already in place in many countries for with substantial costs involved in environmental
plants and plant material, for agricultural reasons. improvements, information and normative statements
Solving problems after they have occurred often seems will not suffice. If binding instruments lose their
difficult. Disease control by vaccination is one activity importance in limiting options, only financial instru-
that is obviously to be stimulated in terms of research ments and liability instruments can substantially cor-
and instruments for their operational application to rect the pay-offs for those making choices, be they
large parts of the population. This level of problem governments, business and environmental associa-
analysis cannot yet be generally related to policy instru- tions, or private producers and consumers. Softer cul-
ments, thus allowing the problem to grow. This is an tural instruments are important to support prime
undesirable state of affairs for broad sustainability rea- movers and to generate political support.
sons. Ultimately, the influence of human activities may Thus, a major problem for future environmental pol-
reach a point where viable ecosystems can no longer icy relates to methods compensating for the diminishing
exist unless fostered by specific human activity. Such a role of technology-specific binding regulations. With
total control of our biotic environment, and hence also few exceptions, the role of financial instruments has so
of our a-biotic environment, is not feasible at a scien- far been limited. Broader use will be based on a number
tific level, let alone that policy instruments could of conditions. To begin with, a clear choice is required
already be envisaged. between issuing emission permits, against the polluter-
pays principle, or creating emission taxes, in line with it.
Secondly, the operational applicability at the level of
emissions and resource use (e.g. CO2 taxes), as opposed
13.2.5 Prospects for Instruments to application at product and technology levels, has to be
improved. Thirdly, international coordination in terms of
As explained above, the role of technology binding the design and levels of taxes is needed to avoid unfair
instruments is expected to decrease further. Only in and overall costly shifts in competitiveness between
industries not exposed to international competition, firms in different countries. Fourthly, better integration
like building and infrastructure, and in the context of of various environmental aspects is required for cultural
some internationally binding agreements on prohibit- instruments, as real actions always involve virtually all
ing emissions like the Montreal Protocol for most environmental problems that exist. This is not only a
ozone depleting substances, may instruments binding matter of information but of clearer normative guidance
private addressees remain dominant (see e.g. Castells as to the relative importance of various environmental
Cabré 1999). International coordination, for instance interventions, based on their potential consequences.
by technology guidelines or Best Available Technology Lack of conceptual clarity is a major problem here. Can
rules as in the EU IPPC Directive (Integrated Pollution we realistically specify the ultimate consequences of
Prevention and Control), may help to leave some CO2 emissions at an ‘endpoint’ level? Or should we eval-
space for technology binding instruments, especially uate at the ‘midpoint’ level of global warming potentials?
in matured industries with relatively little technology Or are potential instability and uncertainty about possi-
development. In other fields, different types of instru- ble catastrophes the prime motives for reducing green-
ment will take over or at least become more impor- house gas emissions? Such questions need to be answered
tant. Private choices in production and consumption in a very practical manner if a trade-off between, for
are increasingly being guided by cultural and infor- instance, energy use, land use and a diverse array of
mational instruments like life cycle assessment (LCA) emissions is at stake, as is the case in most practical deci-
and Environmental Audits, both standardised by ISO. sions in production and consumption. Such normative
13 Environmental Policy Instruments 251

guidance, which is only partly based on empirical evi- already been taken, with governments or private
dence, is not only a prerequisite for practical decisions organisations owning the species on their territories
on specific technologies, but also for more aggregate and their genes, whether naturally found in organisms
developments, as in steering technology development or constructed. This may help to protect and regulate
and creating more sustainable lifestyles. the existence of gene pools. It does, however, not nec-
What is ultimately needed is not only a view on the essarily help to create healthy ecological conditions.
relations between the various environmental effects The preponderance of single protected genes, as in
involved, and the trade-offs between environmental inter- agricultural production, may even lead to monocul-
ventions implied in our actions. The trade-off between tures of such economically dominant genes, with
economy and the sum total of environmental effects also probably negative overall ecological effects through
needs to be stated as a clear principle, guiding the quanti- reduced biodiversity.
fication of all types of instruments in a uniform way. Two lines of instrumentation are presently available
Without such guidance, equal treatment of similar cases for saving and creating ecological values. One is the
cannot be achieved, leading to both injustice and substan- creation of nature reserves, as an option-creating
tial static and especially dynamic inefficiencies. instrument. However, increased spatial needs for food
Finally, structural instruments, such as changes in production to feed the growing and more affluent pop-
institutions, constitute an option for new policy ulations and increased mono-cultural production of
instrumentation that has not yet been fully exploited. biomass for energy purposes literally leave less room
For example, cases against the tobacco industry in for this instrument, given the largely fixed area of land
the US have shown how large the payments can be in on earth (which even will decrease due to rising sea
a specific judicial setting. In all cases, creating the levels). The other, ‘integrational’, line is to create more
right incentives should be accompanied by removing ecological quality in areas primarily used for agricul-
the wrong incentives. The situation in many coun- ture, recreation, infrastructure, production and hous-
tries is such that there are substantial subsidies on ing. Again, the options have hardly been investigated
energy use, as in Eastern European countries, and and instrumentation is largely lacking. Apart from
similarly by tax exemptions as for aircraft fuel on emission reducing instruments, there is a clear lack of
international flights; that the cost of infrastructure is instruments safeguarding ecological richness in diver-
not reflected in prices, as with un-priced roads; that sity and volume, not only in nature reserves but also in
new technologies are difficult to implement due to human-dominated ecosystems. Furthermore, ecology-
complexities of regulation; and that a global perspec- oriented instruments still lack theoretical foundations
tive on environmental policy development is pre- and operational development.
vented, as implied in emerging WTO regulations. If
a broad shift in policy implementation were to
emerge, the result would be a more balanced inter-
nalisation of environmental consequences, both in
13.2.6 Strategic Instrument Choices
public and private actions.
Instrumentation for newly emerging environmental
Ahead
problems is usually lacking. The appropriation of
nature, that is, a decreasing share of nature in the total Our first assumption here is that it is not possible to
biomass production, is accompanied by an even faster avoid choices regarding environmental policy instru-
reduction in natural biomass breakdown, which is the ments by doing nothing. Economic growth, population
source of food for all fungi and animals. This is one of growth and globalisation tendencies make internation-
the new environmental problems, defined here at ally coordinated policy development inevitable. At
‘midpoint’ level, relating to biodiversity, the life sup- the same time, it is impossible to make independent
port function of ecology and the quality of nature. choices on policy instrumentation for each individual
Instruments that can work at a global level hardly case. A well-argued strategy has to guide serious
seem to be available. For some aspects, such as the development of policy instruments. If the polluter-pays
protection of available genetic information, the road principle is accepted, all instruments have to take it
towards structural-institutional instruments has into account. Basic choices should preferably be made
252 G. Huppes and U.E. Simonis

consistently, according to generally recognised princi- to operate an installation. If the number of permits
ples. Some major lines of development are discussed brought on the market by governments is set so as to
here, relating to basic liability rules, environmental achieve a predetermined price level, the difference with
ethics, the importance of efficiency and equity, means- an emission tax of a similar level is really very small in
directed or goal-directed types of instruments and, terms of effects, and may even be similar in its admin-
finally, principles of policy integration. istrative operation.
At a practical level, the liability instrument works
by linking cost to environmental interventions. The US
[Link] Liability Rules has taken a practical approach in the field of waste
management, by making firms liable for the sanitation
Liability rules have traditionally been set up to prevent cost of illegal landfill with chemicals. This programme
active infringements on the goods or rights owned by (Superfund Programme 1986) has conferred substan-
others, either individually or collectively. The 1960s tial cost on polluters, based on ongoing sanitation
saw the start of a debate in economic circles on the activities pre-financed by government, see e.g. EPA
other option, that is, giving everybody the right to (1999). Similar programmes have been developed in
infringement of the goods or rights of others, especially the Netherlands. In the US, these programmes function
in the environmental domain. The discussion was against a background of already existing extended lia-
opened by Coase (1960), who showed that there was no bility (‘chained and several liability’), where larger
difference between these options for the outcome in firms in the chain are liable, and if the damages result
real terms, if transaction costs needed to arrive at these from emissions of several firms, each one is liable. In
outcomes could be neglected in both cases. In the Europe, developments towards extended environmen-
debate following his paper, the latter restriction has tal liability are ongoing, based on the Environmental
been somewhat relaxed in that Coase’s conclusion also Liability Directive.
holds if transaction costs are similar. Transaction costs
result from the necessary contacts, primarily the labour
time of all involved, including costs of negotiation, [Link] Ethical Norms
control and litigation.
This point of view has had a major influence on envi- In the international setting of a globalising world, basic
ronmental policy, where the same period had seen broad discussions relate to the ethical principles which pro-
acceptance of the polluter-pays principle. Different ver- vide guidance in handling distributional effects. Though
sions of this principle exist. A basic element is that pol- sustainability involves an accepted distributional prin-
luters have to pay for the environmental damage they ciple, concrete ethical norms have not been agreed
cause, thus internalising environmental aspects in their upon. Should all citizens of the world have an equal
decision-making. Since Coase showed that the principle share in the environmental use space? Is this share trad-
is not required for cost-effective policies, policy-makers able? Does every citizen have the same right to a cer-
may reason towards policy instruments on the basis of tain minimum environmental quality? Are the costs of
net costs in terms of real outcomes including transaction environmental improvements to be distributed equally
costs, and do ‘what is best’, on a case-by-case basis. per head of the population? Or is an equal percentage of
This has opened up the path to tradable emission per- income to be spent on environmental protection?
mits, which give the owners a direct right to pollute. The answers to such questions have a direct bearing
The conflict between the polluter-pays principle and on instrument choice. Internationally, the use of emis-
the pragmatic do-what-is-best in each case principle sion rights as a political-administrative instrument, with
somehow needs to be clarified and resolved to allow initial rights distributed to countries according to their
basic innovation in national, and particularly in interna- share in world population, would be in line with equal
tional policy instrumentation. In-between options are environmental use space. Emission permits inversely
possible, but not necessarily more attractive (see Tisdell related to average income give more per capita emission
1998). One option is to market emission permits with a use space to the poor. Equal emission taxes worldwide
very limited term of validity. Effectively, this means would roughly lead to equal shares of income being
that the right to emit has to be bought again each time paid for environmental protection. If such principles are
13 Environmental Policy Instruments 253

to be more than a guise for tactical interest protection, a problem in this approach. This means-oriented
i.e. if they have real meaning, the implications for policy approach is contrasted with the goal-oriented approach,
instrumentation will be quite direct. Still, there is not in which policy instruments have to internalise sustain-
one principle that can force the decision. ability goals as fully and directly as possible, allowing
for decentralised technology choices with incentives
for environmental improvements.
[Link] Efficiency and Equity It is clear that efficiency considerations also imply a
choice to adopt the goal-oriented option. In the liberal
A further strategic choice concerns the relation between ideal, the choice is clearly to adopt goal-directed instru-
global efficiency and global equity. For the sake of effi- mentation, while in social-democratic and socialist cir-
ciency, marginal costs of environmental protection or cles, preferences might lean more towards means-directed
improvement should be equal across the entire world, instrumentation. However, ideas in European social-
which means that environmental improvements should democratic parties indicate that the broad integration of
be achieved in all choices, up to a particular level of cost environmental considerations in private decision-making
per unit of improvement. If this rule is not satisfied, with is to be preferred over the option of having governments
some doing less and others doing more, the world could decide on technology choices on a case-by-case basis.
benefit from a shift in effort, with those still having cheap Thus, there is a broadly supported political tendency
options for improvement doing more and those with high favouring a shift towards goal-oriented instruments,
costs of improvement doing less. A real Pareto improve- though means-oriented instruments remain the main
ment is then possible, with everybody being better off if vehicle for environmental policy, also because the con-
those who reduce efforts compensate for those increas- trols for regulators seem more easy to implement.
ing their efforts for environmental improvements.
The broad emphasis on efficiency as a guiding prin-
ciple for trade relations indicates that this principle would [Link] Principles of Policy Integration
also have prime importance in environmental policy
instrumentation. Internationally tradable private emis- In means-oriented policy instrumentation, the integra-
sion permits and emission taxes that are the same the tion of the various environmental aspects involved is
world over would be prime instruments. Who is receiv- implicit. One may assume, optimistically, that a single
ing the ‘grandfathering’ rewards of initial permit distri- policy-maker will be consistent in the way he or she
bution and who is receiving the proceeds from the uses the implicit trade-offs between various environ-
emission taxes is not relevant to efficiency consider- mental aspects, as well as the trade-offs against social
ations. This indicates that there is some room for com- and economic aspects. If multiple policy-makers are
bining efficiency with equity, by redistributing proceeds. involved, in both public and private organisations, con-
Full emphasis on equity will lead to other instruments, sistency cannot be expected to come about automati-
however. The justice principle (as embodied in the pol- cally. In goal-oriented policies, there is a logic to
luter-pays principle) would shift the choice from emis- making an explicit statement on the relative importance
sion permits to taxes on adverse environmental impacts. of various environmental aspects related to activities, at
the operational level of emissions, extractions and dis-
turbances, that is, trade-offs are to be stated more
[Link] Goal-Oriented and Means-Oriented explicitly, allowing for equal trade-offs in different sit-
Instruments uations. An example is the equivalence between 14 t of
SO2 and 1 t of CO2 which is being used informally in
A further major choice concerns the aim that has been Dutch environmental policy. Worldwide, however,
made operational in the environmental policy instru- explicit statements on such trade-offs are lacking, while
ments. For instance, an emphasis on easy implementa- decentralised environmental decision-making cannot
tion means that regulators have a clear grip on do without them. Where tradable emission permits
technology development and makes policy integration come up, the comparative reference may be the price
an aspect of policy development. Dynamic efficiency, paid for the permit. This may be the case in the European
the most important cost aspect in the long run, remains Union for climate changing emissions (in the Emissions
254 G. Huppes and U.E. Simonis

Trading Scheme) if it develops to a mature system, and The main focus here will be on public policy, but the
in the US for acidifying emissions. These markets are definition is not limited to this. The internal tradable
very partial and not yet global however. permit system developed by Shell (see Box 13.4 in
Reasoned choices in this respect require explicit state- Section 13.4) is a perfect example of an environmental
ments about the reasons for choosing particular trade-offs policy instrument. It has been superseded by public car-
and the links between these environmental aspects and bon trading systems like in Norway and the EU, see
concrete economic actions. Such relations depend on Hoffman (2006). Of course, not all policy instruments
evolving normative ideas about what is important, on the are intended for environmental policy. Other instru-
state and development of the environment and society, ments for public policy, like those on energy and
and on the way these relations can be modelled. transport, may include environmental policy goals in
Consistency, including consistency over time, can only addition to the prime non-environmental goal, or instru-
come about on the basis of an explicit and comprehensive ments may be multi-goal oriented, or the goals are not
debate. This ideological superstructure to operational explicit, as with excises on petrol and alcohol. Multiple
policy is poorly developed. In Dutch environmental pol- goals are implied in integrated policies. The division
icy, the ‘themes approach’ was developed as a conceptual between environmental and non-environmental instru-
framework over a decade ago, including themes like ments is thus not strict. This is not a real problem, how-
eutrophication, acidification and ozone depletion. This ever, as policy instruments for non-environmental goals
approach has been followed in various ways by other may be analysed in a very similar way.
countries. The European Union, for instance, has defined Not all policies are structured, in the sense of being
a large number of Preferential European Environmental institutionalised in terms of instruments. Setting up the
Problems, not as a systematic treatment of the subject but high-speed rail link to reduce air traffic between Paris
based on political consensus, including quite incommen- and Lyon did indeed reduce air traffic at first, and reduced
surate items ranging from waste prevention to biodiver- its growth afterwards. Green politicians may exhort the
sity preservation. Waste prevention, however, is not an people in public speeches to leave their cars at home for
environmental aim but a means to achieve environmental at least 1 day a week, with some success. Such incidental
aims. This means that the means-oriented approach is activities towards policy goals, however, are not seen as
reintroduced through the backdoor. The explicit and gen- instruments, as instruments would need to have an ele-
eral normative integration of environmental policy aims, ment of repetition in their application and operation.
as opposed to the implicit choices sufficing for technol- Hence, if high-speed railroads are built consistently on
ogy binding instruments, is a clear task ahead. trajectories with rising air traffic, this provision of infra-
structure can be regarded as an option-creating type of
policy instrument. If a politician’s speeches are part of a
series set up for public education, they too may be seen
13.3 Policy Instruments: What Are as part of a communicative instrument. The dividing line
They Good for? is not strict, but, again, this is not really a problem.
As a final distinction in the definition, it may not be
clear what exactly the environmental policy goals of
13.3.1 Policy Instruments: What Are some instruments are, and if these are really environ-
They? mental goals. Raising prices for dumping waste in
landfill sites may have non-environmental aims, for
instance to increase the availability of easily accessible
There are many instruments that might be relevant to
landfill sites, or to provide an incentive for increased
environmental policy, including analysis tools, check-
use of under-utilised incineration plants. Or it may be
lists and plans. More generally, environmental policy
seen as a means of reducing primary materials produc-
instruments can be seen as the means of implementing
tion, reducing resource depletion and the environmen-
such policies. Here, a slightly more restrictive defini-
tal effects related to ores processing. What exactly
tion is used:
constitutes the ‘real’ prime motive is often difficult to
Environmental policy instruments are structured activities aimed
at changing other activities in society to achieve environmental establish, but also not very relevant. Such borderline
goals in a particular time schedule. instruments may still be analysed as instruments for
13 Environmental Policy Instruments 255

environmental policy, with environmental effective- evaluative analysis may be part of the policy cycle, in
ness as one aspect of their assessment. The actual envi- which the effects of policy, combined with other devel-
ronmental effectiveness of instruments may not always opments, feeds back into new policy preparation. The
be a distinguishing criterion. In certain circumstances, evaluative part is worked out in Section 13.3.6 below.
subsidies on environmental improvements, for instance, The framework for the empirical part of the analysis
may have adverse effects, by delaying structural has four main components: regulators, regulatees,
change which otherwise would have taken place. These society and the environment. These four components,
subsidies are then environmental policy instruments as applied to one particular country or region, are mir-
that are not adequate for the particular situation, but rored by the same entities abroad (see Fig. 13.3). The
they still are environmental policy instruments. framework defines the basic structure for modelling
the functioning of environmental policy instruments.
In a very basic mechanistic model, there is a single
causal chain from the regulator’s actions to the envi-
13.3.2 Why Have Policy Instruments? ronmental effects. This limited framework already
opens up a world with a rich variety of instruments and
Why bother about environmental policy instruments a high level of complexity of mechanisms in society.
when actual policies based on integrated assessments The starting point in the model is some public regu-
can integrate environmental and other consequences in lation, which may be collective, as when a public body
everyday actions? The main reason is that the complex- sets an emission standard in the metals plating industry,
ity of all empirical relations and that of assessment are so or private, as when the management board of a firm sets
great, and the information requirements so vast, that this an environmental performance target. As a first step in
option is not really available beyond a limited number of the causal chain, there is the technical adjustment
relatively simple occasions. Instruments work by simpli- enforced among regulatees as the subjects of instru-
fying a complex reality. They can be studied and assessed ment application, see step 1 in Fig. 13.3. A second step
at a general level to decide what conditions can be identi- usually centres around economic mechanisms and
fied for their sensible application. This may reduce the
complexity of policy-making. At the receiving end, in
society, most policies have their effects not in terms of national foreign
directly correcting current activities of regulatees but, to 6
policy policy
a large extent, by guiding the planning of and decision- makers makers
making on future activities. Having instruments whose
nature is known from the relevant literature and past 1
experience will make policies more predictable and
facilitate adjustments. These adaptive mechanisms in direct regulatees
society, if structured in stable patterns, can be regarded regulatees abroad
as a part of policy instrumentation as such, and may be
more important than specific policy applications them- 2 5
selves. Without such applications, however, the general
national other
adaptive mechanism would cease to exist. society societies

4
3
13.3.3 A Framework for Analysing
Policy Instruments environment
national abroad /
environment global
Various policy instruments may be characterised in a
common framework, with an empirical part explaining
how they work and an evaluative part providing crite-
ria on their value and adequacy. Such an empirical and Fig. 13.3 Regulation: a simple model without feedback loops
256 G. Huppes and U.E. Simonis

relates to the costs imposed on regulatees, that is step 2 and technology adaptations. For instance, costly mea-
in Fig. 13.3. The degree to which such secondary effects sures to reduce emissions in the metals plating industry
are taken into account may vary. Effects on markets and have induced a shift to high quality coatings, with other
other technologies will usually be part of the analysis, types of emissions resulting and with other policies
and will depend on the specific circumstances in these being required. On the other hand, inducing changes in
markets. Stricter emission standards in a small, open an industry may mean that cost-saving innovations that
economy with a few large internationally operating were already available come to be introduced faster and
firms may lead to emission reduction by shifting pro- on a wider scale.
duction to locations abroad, without necessarily chang- The ultimate feedback, of course, is that relating to
ing technologies, and without reducing emissions at a environmental quality. The poor air quality in Mexico
global level. Conversely, technology adjustments in a City not only raises direct costs of production by
large country with many small firms producing for the requiring air filters in many processes, it also reduces
national market will be more pronounced, with only the legitimacy of the government and makes it difficult
limited changes in the volumes produced. For a given for firms to attract managers and specialists from
set of national technical effects and volume changes, abroad. Visible actions, in terms of standards and regu-
the net resulting environmental interventions can be lations, are most suitable to remedy these negative
defined and linked to effects on the national environ- effects in the short term, by showing that ‘something is
ment (3) as well as on the environment abroad and ulti- being done’. Less visible actions, like changes in lia-
mately the global environment (4). As most markets are bility rules and market structure, might, however, be
now international, national policies will induce eco- more effective in the long run. Hence, various feed-
nomic changes abroad (5), which also have certain back loops can influence instrument choice. Feedback
environmental effects. Finally, policies in one country mechanisms by their very nature are dynamic, with
may directly influence policies abroad (6). Dutch past choices determining future ones. Once a particu-
excises on petrol, for instance, are limited by the lar policy for some environmental problem has been
German excises on petrol, as too large a difference will designed, for instance one issuing emission rights, it
lead to the closing of petrol stations in the border regions. becomes very difficult socially and even juridically to
Similarly, Californian regulations on ‘emission-free cars’ change over to policy instruments that are more in line
have set in motion regulatory activities and technology with the polluter-pays principle, where emitters have
development in Japan and Europe. to pay for their infringement of the right of others not
The model with one-way causalities does not, however, to be polluted.
correspond fully to reality, where feedback mechanisms, Why is the question how instruments function so
always dynamic, abound. Such feedback mechanisms important for their characterisation? The answer is that
may be quite complex. If regulatory capacity is limited, as in reality, instruments are not independent given
in some way it always is, using regulatory power to ‘things’; their definition and description, and the anal-
solve one problem precludes its application for solving ysis of their functioning, are intertwined and closely
others. Also, using one type of instrument for one prob- related. Most instrument definitions focus on their
lem, like covenants to achieve the best available practice functioning, mostly on only one aspect of it. Thus,
for energy saving in industry, will render the later intro- technical prescriptions focus on technologies applied
duction of emission taxes on CO2 and NOx rather unac- in industry; tradable emission permits focus on the
ceptable to industry. Negotiations on a covenant depend equalisation of marginal emission reduction costs
on what the industry sees as an alternative to the cove- between firms, industries, or countries; liability rules
nant: maybe emissions taxes or maybe costly actions. focus on specific enforcement procedures and actual
Hence, such negotiations necessarily take place ‘in the compensation and covenants focus on the procedure in
shadow of the law’, as Galanter (1981) and Scharpf the policy formation process. None of these descrip-
(1991) phrased it. Indirect effects in society, in terms tions take all steps in the framework into account, let
of the economic and environmental developments they alone the feedback loops that usually exist.
induce, result from further feedback mechanisms. Limited description may easily lead to simple
A very common mechanism is that by which regula- assumptions about the other steps in the causal chains
tions induce costs and hence lead to market changes required for environmental effectiveness. For instance,
13 Environmental Policy Instruments 257

many believe that emission permits may not be ideal in no behavioural consequences in the firm itself (see
terms of minimising costs, but that at least they are a Cole and Clark 1998) (Box 13.2). By contrast, similar
sure means of reaching specified results. In most coun- taxes in capitalist countries with competitive markets
tries, however, this belief is not well founded (Bardach may induce far-reaching behavioural changes. For
and Kagan 1982; Bonus 1998; Hawkins 1984; Vogel instance, Dutch wastewater taxes enacted in the 1970s
1986). Rules often exist on paper only and are not nec- were followed by overall decreases in effluent volume
essarily linked to actual practice, though the tendency by a factor of 20, mainly through process-integrated
towards formalisation as taking place in Europe may technology changes (Bressers 1988; Huppes and
lead to closer correspondence. While environmental Kagan 1989). Endres (1997) has described the contex-
standards and regulations in the former Soviet Union tual requirements for an effective application of mar-
were among the most stringent in the world, environ- ket-based instruments.
mental quality was worse than in most other countries. In some Western countries, like England and the
This means that defining rules and regulations, or at Netherlands, policy development and implementation
least viewing their functioning in a broader framework, have been linked in a less recognisable way. It was
helps to avoid the myopia of partial views, making the broadly accepted that firms functioned without the
context of their functioning more important. obligatory permits (see Vogel 1986). In such a ‘slightly
illegal’ situation, regulators might actually have greater
influence on developments than if they had issued
seemingly strict permits which tended to petrify the
13.3.4 Policy Instruments in Context past. Tendencies towards a more formalised and liti-
gious type of society, as in the US, have made this style
Most people would agree that policy instruments should of regulation more difficult. In the old situation, there
be placed in the broader framework of their function-
ing. However, if this functioning were the basis of their
definition, this could lead to counterintuitive results. Box 13.2 Main lines of argumentation in Section
Specifying this framework may show that what is 13.3
referred to as one instrument is actually something dif-
• The prime role of policy instruments is in reducing
ferent in different contexts. And the implementation of
social complexity to manageable proportions.
one and the same instrument may be very different in
• Instruments as institutional arrangements may
different prevailing circumstances. In litigious societies
not only be seen as tools used by governments to
with limited general legitimacy, technology-binding
influence private behaviour, but also as a means
legislation may be implemented effectively only by
of guiding behavioural relations between public
various types of lawsuit, with years of delay, while in
bodies, and between individuals and private
highly integrated and less formalistic countries, legisla-
organisations.
tion and implementation may be nearly synchronous,
• Environmental problems mainly result from the
seemingly originating from informal consensual com-
external effects economic activities have on col-
munication (see Vogel 1986). Conversely, for regula-
lective goods.
tees, technology-binding legislation may lead to adverse
• Environmental policy instruments help to avoid the
reactions or to internalisation of the rules enacted. Are
tragedy of the commons by solving the prisoner’s
these two separate instruments or are they context-
dilemma and preventing the free rider problem.
related applications of the same instrument?
• The evaluation of instruments for environmental
The broader effects of policy instruments in society
policy is not only based on first-order criteria for
depend greatly on previously established institutions.
the evaluation of effects, like eco-efficiency and
Although several communist countries have enacted
distributive justice, but also involves second-order
emission taxes, they have had little effect. As increases
criteria, like effects on competitiveness and influ-
in production volume were the primary aim of state-
ence on technology development, and third-order
owned firms, with prices fixed and with the state reap-
strategic criteria, like fitting in with general insti-
ing the profits and shouldering the losses, emission
tutional, cultural and economic developments.
taxes were simply entered on the balance sheet, with
258 G. Huppes and U.E. Simonis

was bargaining in the shadow of the (possibly unrea- reality and make the behaviour of all concerned more
sonable) law, with a permit as the eventual short-term predictable. If the instruments in official use hide what
fixation of a situation. In the new, more formalised isactually happening, they increase complexity and
situation, it is not so clear how administrators can have may more usefully be removed. In administrative sci-
a flexible influence. The covenant has taken over the ence and in sociology of law, the consequence has

bargaining step, while the shadow has not been clearly been that the idea of standard instruments has more or

defined. Thus, the precise definition of an instrument less been abandoned. In an ideal horizontal govern-

depends on a more accurate assessment of its function- ment, all stakeholders participate, in principle on an
ing. In the Dutch and English contexts as they were, equal footing, resulting in deals that are best suited to
permits were actually not the 'real' instruments for the situation (see von Benda-Beckmann and Hoekema
environmental policy at all; rather, they formed the 1987). Again, why then bother about instruments? The
background for negotiations, with mostly informal answer is not straightforward. Governments may
deals between regulators and regulatees achieving the increasingly stimulate decentralised use of analytical
actual environmental improvements (Box 13.3). tools like LCA to guide the outcomes of negotiating
Recognition of the contextual specificity of policy procedures in the right environmental direction. Such
instruments seems to be detrimental to the basic aim of sensible developments should not be denied when dis-
distinguishing policy instruments, which is to simplify cussing the role of instruments; they can instead be

Box 13.3 Sectoral emission reduction covenant principle on LCA inventory modelling of the activ-
A covenant between the Dutch central government ity with and without the measure.

and NOGEPA (Netherlands Oil and Gas Exploration The further effects on the environment are speci-
and Production Association) specifies goals for fied in terms of policy themes (see Guinée et al.).
emissions reduction and the procedures to be used to Eco-efficiency is established by first aggregating
achieve them. These goals relate to the performance these theme scores into one eco-indicator score,
of the firms involved, not individually, but as a using covenant-specific weights established by all
group. The means used to meet the emissions reduc- those concerned in the covenant (see Huppes et al.
tion goals, like technical measures, are not specified 1997). This is one of the few examples worldwide,
in the Covenant. The procedures to meet the goals in a regulatory context, where trade-offs between
are specified, however, including procedures to environmental aspects are specified explicitly.
redefine the goals when circumstances require this, Combining it with cost allows the environmental
for instance because of technical and economic cri- improvements per Euro of expenditure, that is, the
teria. For the actual implementation, all firms cost-effectiveness, to be established for each emis-
involved are to specify an Environmental Business sion-reducing technique in its specific context. The
Plan (BMP). Collectively, the result of the efforts firms involved can make an inventory of possible
indicated constitutes the Industry Environmental techniques and rank them according to this cost-
Plan, leading to the goals specified. effectiveness. The most attractive techniques in this
There is no strict relation between the overall respect are included inthe four-yearly Environmental
emission reduction targets and the contribution dif- Business Plan. Plotting this against the cost per unit

ferent firms have to make to this target, which is at of environmental improvement results in a flat
least partly a subject for collective consideration. curve, which at a certain point starts to rise steeply.
The Covenant stipulates that specific measures are It is in the interest of the firms involved to develop
to be selected based on their cost-effectiveness, or technologies in the flat range, to avoid high costs in

eco-efficiency, with specific rules to quantify this the next Environmental Business Plan.
efficiency. The usual private costing methods are For a general description of the covenant, see

applied for the economic analysis. The environmen- [Link], section Environment. The full text
tal analysis starts with the life cycle environmental of the Covenant (in Dutch only) is available at the

interventions for each technical measure, based in same website (Last visited: August 2007).
13 Environmental Policy Instruments 259

made part of the development of environmental policy stench and emissions? One option here would be to
instruments. Therefore, instruments can also cover limit the growth of airports, which would not require
situations where governments may be quite invisible specific instruments for environmental policy, as long
or even absent. as building airports requires public decision making.
Thus, the question is not really why we need envi- Another option, however, would be to use the price
ronmental policy instruments but which ones we need, mechanism for environmental purposes. Taxing the
or are able to have. There are several reasons why the emissions of carbon dioxide, nitrogen oxides and noise
decentralised horizontal instruments do not suffice. at sufficient rates will reduce these emissions, not only
Firstly, there are limits to horizontal government, in through technology adjustment but also through a
terms of the human resources and knowledge required reduction of the numbers of passengers transported.
for adequate negotiations. This limitation exists on the The growth of airports would then be reduced as a con-
side of government but also on that of other stakehold- sequence of environmental measures, not as an indirect
ers. Most firms hate continuous negotiation because it means of fostering the environmental aim of emission
drains their management capacity and so endangers reduction. Using instruments for environmental policy
their current and future functioning. Sustainability in this way may prove to be more useful than seemingly
requires continuous adjustment of the behaviour of all simple measures like preventing airports from growing.
firms and all consumers, which is now mainly guided
by market considerations. Influencing this behaviour is
clearly beyond the scope of the negotiating govern-
13.3.5 Environmental Problems: Causes
ment. Hence, corrections to the outcomes of in-firm
decision-making, including technology development and Solutions
and product design, and of market processes can be the
subject of negotiating governance only in special situ- Understanding the working mechanisms of instru-
ations, within the capacity limits of regulatory bodies. ments in solving environmental problems requires
In addition, there is a more positive reason for having some insight into the causative mechanisms resulting
environmental policy instruments. Institutional devel- in such environmental problems. In virtually all causal
opment in society somehow has to cope with sustain- models for societal actions, some rational actor models
ability in a structural manner. Leaving a key value like play a key role. It is in such rational actor models that
sustainability to day-to-day negotiations by private many of the causes of environmental problems can be
persons and lower level officials would be unwise or identified and specified, at least for a start.
even immoral. Somehow, quasi-automatic mecha- Common to all environmental problems is the
nisms, such as institutions, have to be shaped to safe- causal mechanism by which direct private advantages
guard the sustainability of operational, tactical and of some actions outweigh the adverse environmental
strategic decisions. Environmental policy instruments effects for the persons (or organisations) deciding on
that do not depend on horizontal negotiations will play that action, while at the same time this adverse effect
an indispensable part in these mechanisms. In between, may affect others. To this extent, the adverse effect is
there are the horizontal negotiations to decide which external to these private considerations: it is an exter-
instruments to apply and how to apply them in practi- nal effect of private actions on some large collectivity.
cal policies. If such negotiations are to succeed, other If the single owner of an island cuts down his forest to
instruments must be available as well; these are the create a garden, he simply prefers the garden to the
‘harsh’ instruments, serving as more or less ready forest, and there is as yet no environmental problem
options for governments to base their negotiating posi- involved. It is only if others are bothered by the disap-
tion on. They constitute the ‘shadow of the law’ in pearance of the forest or the consequences of its disap-
which governments can safeguard the sustainability of pearance that there is an environmental problem. This
the outcome of negotiations in the networks involved. shows the collective good nature of environmental
There remains a domain where instruments at first quality, and is a necessary mechanism for environmental
sight may not seem to be relevant, as with some single problems to occur. In economic jargon, it is external
big issues. For instance, should we not just curb the effects of private actions that are detrimental to a
further growth of passenger air transport with its noise, collective good.
260 G. Huppes and U.E. Simonis

An additional mechanism in problem development rationality are relaxed to include such social aspects
is that the detrimental effects usually result not from of human behaviour, the unpleasant situation remains
the actions of a single person but from the combined that detrimental environmental effects do occur and
actions of many persons. Though not strictly necessary are re-created all the time by people who know the
in a logical sense, this is the typical situation for nearly problems exist. So, even after taking into account the
all environmental problems. It is the tragedy of the social nature of behaviour, the wrong choices are still
commons. Single actions may contribute very little to made so often that environmental problems result.
the problem, but it is the multitude that leads to overall On the basis of these theoretical considerations,
undesirable effects, and ultimately to the breakdown of sometimes named the ‘field model’, it is now possible
the ecological system. to specify what environmental policy instruments
Together, these two mechanisms have a power that should do. They should:
is hard to break. If an individual producer or consumer
1. Avoid external effects on the environment and thus
corrects his behaviour, his action may have negative
save collective goods, in this case related to envi-
effects on himself, in terms of costs and efforts, even if
ronmental quality
only in terms of the burden of nuisance. At the same
2. Avoid the tragedy of the commons
time, the positive environmental effects of the adjusted
3. Solve the prisoner’s dilemma
behaviour may well be negligible. Such cases result in
4. Prevent the free rider problem
the prisoner’s dilemma. Rational actors will only
choose the behaviour with the preferred outcome if The model also allows us to indicate the mechanisms that
they can expect most others to act in the same way. In policy instruments may be aimed at. The behaviour of an
the ‘wrong’ situation, actual behaviour by others individual actor can be corrected in a number of ways
proves that this expectation is not justified and rational (see Bressers and Klok 1988 for a fuller treatment).
actors will choose the sub-optimal behaviour of con-
1. The set of available alternatives can be changed.
tributing their share to the environmental problem. If
corrective action is taken by all but a few, the environ- This can be done by offering new alternatives, like
mental problem is largely solved, benefiting these few separate collection facilities; by removing alternatives
while they do not bear their share of the costs. This is physically, as by fencing conservation areas; or by
the free rider problem. If the free riders are visible, improving knowledge of existing but as yet unknown
social norms on collective action may easily become alternatives, as in nature education programmes.
eroded, and with it the collective environmental good.
2. The consequences of alternatives can be changed.
If the free riders are invisible, the flesh is weak. It then
takes highly internalised values for most people to This may be done positively, as by subsidies on unleaded
remain on the right track. Environmental policy instru- petrol, or negatively, as by threatening those who dump
ments somehow have to break this circle, either by toxic wastes with jail sentences or other penalties.
effectively influencing all (or almost all) people, or at
3. The evaluation of the consequences of alternatives
least by creating the confidence that most others will
can be changed.
act accordingly.
A criticism of this model has been that the rational This can be done by changing the value system of
actor model underlying it is based on too simple a actors, through educational processes, or by improving
view of the real motives of real people. In reality, their active knowledge of the consequences of available
many actors indeed often behave altruistically, because alternatives, as by eco-labelling schemes.
they like doing so, and groups of actors often have an These three types of mechanism in a policy instru-
explicit or tacit mutual understanding to avoid ‘bad’ ment may avoid the external effects on the collective
behaviour (see Sen 1977). At least some people bother good, as well as the prisoner’s dilemma and the free
about separate collection of waste streams, even in rider problem. In many situations, however, this is pos-
instances where others cannot see what exactly they sible to a limited extent only and the situation may still
are doing. They have internalised environmental norms be that of a prisoner’s dilemma, with the free riding
to some extent. However, even if the restrictions on option lurking in the background. If only heavy industries
13 Environmental Policy Instruments 261

were induced to reduce their CO2 emissions substan- 13.3.6 Evaluation Criteria for Policy
tially, using a large number of specific measures, the Instruments
prisoner’s dilemma would remain for all other actions
not regulated, for which free riding may still remain the
norm. In addition, the non-heavy industries would still The analysis of how instruments for environmental
be riding free. Not only would they continue to emit, policy work is one part of policy instrumentation hat is
but lower energy prices because of reduced demand by indispensable for any evaluation. The question is, how-
heavy industry may even allow them to increase their ever, what criteria to use in judging policy instruments.
emissions. Thus, there is another role of policy instru- Somehow, the framework for the evaluation of instru-
ments in avoiding the tragedy of the commons, by solv- ments, and instrument-related policies, is to mirror the
ing the prisoner’s dilemma through creating the trust empirical analysis, which ultimately has to indicate
that everybody will indeed take part in the creation of effects in terms of these evaluation criteria. This adds a
the collective good. layer of analysis of a normative and political nature.
As it is the consequences of instruments which are
4. Justified trust in everybody’s positive and due used as the basis of evaluation, the approach is that of
contribution to our common good can be created. consequentialism, not in the narrow sense of a utilitar-
This would make free riding virtually impossible. In ian type of economism, but in the broad sense given to
this situation, the individual may seem to be deciding this term by Sen (2000). In this broad view, conse-
alone. In fact, however, he is making his decisions as if quences obviously encompass the preferences of indi-
he were the collectivity, deciding for all people simul- viduals, as is the case with utilitarianism exclusively.
taneously. This co-operative solution is a very direct However, collective aspects like ‘sustainability’ may
option for solving environmental problems, with col- well also be covered by this broader consequentialist
lective values being internalised in individual decision- approach. Most of the criteria for evaluation specified
making. Although this option seems highly idealistic, below belong to this second approach (see Table 13.1).
it is a normal solution to many problems, at least in Three groups of criteria are distinguished. First-order
small communities. Most people take the trouble to criteria are related to more or less direct operational
store their waste in waste bins, even in larger commu- consequences of the application of the instrument,
nities. Only in situations of reduced awareness of with environmental effectiveness as the first criterion.
norms, as when alcohol is used collectively, do the Second-order criteria relate to broader aspects of
norms on littering break down. Tasks for the common administration and economy, like administrative capac-
good are executed, that is, behaviour is adjusted, ity and technology development. Strategic criteria, the
because one expects everybody to do so. Still, this most general category, link the instrument to the
ideal is not always achieved, even if only a small num- broader culture and institutions in society.
ber of people are involved, as can be seen in some Sustainability may be an agreed general goal behind
families where children (or indeed parents) may try to all specific environmental policies, as in many countries
avoid the daily dish-washing duties, always with good it already is. Its specification is normative and political.
excuses at hand. While the environmental effectiveness of instruments is
What are the requirements for this type of co-operative an indisputable part of their evaluation, ‘sustainability’
behaviour? One element would be that free riders and ‘environment’ are to some extent catch-all concepts,
undergo negative sanctions for free riding when they without precise underlying definitions.
are caught. This would mean that the bad behaviour is The question is in which terms and with what level
forbidden at the level of the individual, and hence the of detail this environmental effectiveness is to be
co-operative approach is absent. However, control and established empirically, taking into account the mech-
sanctions may be more informal, not involving police anisms and time horizon? Is there a right to some
and administration but friends and family, or the minimum quality everywhere, for which permits
next-door neighbours. Another prerequisite is that might be the most suitable instruments, or is some
the behavioural norm is clear and non-compliance overall level of emission reduction enough, for which
is visible. emission taxes might be an adequate instrument? Are
262 G. Huppes and U.E. Simonis

Table 13.1 Criteria for evaluating policy instruments

First-order criteria Second-order criteria Strategic criteria
• Effectiveness (st, lt)a • Social and political acceptability • Fitting in with the broader conceptual framework
• Social costs (st, lt) • Within administrative capacities for public policy
• Eco-efficiency • Limited changes in competitiveness • Fitting in with the broader institutional
• Distributive justice: • Incentive for sustainable technology framework of society
– Intragenerational development • Fitting in with general cultural developments
– Intergenerational • Fitting in with general economic
– Justice as fairness developments
• Generative equality
a
st: short-term; lt: long-term.

cost-effectiveness and efficiency important parameters preferred option, the Dutch government chose to solve
for assessment, as costs will be important for most this collective action dilemma differently, by making
people? Or are distributional effects more relevant for leaded petrol more expensive through a tax measure,
instrument choice? Are economic and environmental equivalent to taxing lead. Thus, leaded petrol was
effects abroad to be taken into account, or only effectively forced off the market, and everybody paid
national ones? Do other aspects of justice, like the the higher price of unleaded petrol. In this solution,
right to pollute versus the right not to be polluted, play burdens for environmental improvements are shared
a role in instrument choice? Is freedom of choice on equally, in the sense that everybody pays the same
the part of producers and consumers an independent price per litre. This is in line with one of several justice
criterion? Even if one were to refrain from normative criteria, which states that the effort required for a cer-
choices in these respects, the same questions return as tain amount of environmental improvement should be
empirical ones in developing and implementing poli- the same for everybody, at the margin. It is not an equal
cies. Other people will mind distributional effects, effort per head, as those who drive more and drive big-
people abroad do mind being polluted, and a broadly ger cars also pay more. This criterion happens to be
defined polluter-pays principle is generally accepted, nearly equivalent to that of economic efficiency, in its
implying that he who pollutes should be held respon- static variant (Baumol and Oates 1988). This efficiency
sible for the consequences of his actions. If people goal is central in economists approaches to instrument
perceive policies as running counter to their values, design and evaluation, see e.g. the survey paper by
the legitimacy of these instruments is reduced, as is Bohm and Russell (1985).
their effectiveness. No doubt, the normative accept- It would, of course, be strange if criteria for judg-
ability of instruments is one major empirical factor in ing environmental policies were different from those
both their political relevancy and their environmental valid for other policies. Hence, the criteria are related
effectiveness. Thus, the normative questions return to general views on the tasks of government. The sum
through the backdoor. of neo-liberal and social-democratic views roughly
When unleaded petrol came on the market – at a encompasses the entire political spectrum, with dif-
slightly higher price than leaded petrol, accompanied ferent political groups emphasising different aspects,
by a government campaign to ‘buy green petrol’ – but by and large involving the same ingredients.
many people’s reaction was that if they were to buy Giddens has compared the new consensus on public
green, they would be part of a minority shouldering the policy tasks with more traditional views. These new
costs, while the main problem would remain unsolved. views are very much related to structural develop-
Thus, buying green would have a limited or even nega- ments in the economy, with global markets and interna-
tive effectiveness, combined with an unfair sharing of tional networks replacing command and control in
burdens. On the basis of their normative appraisal, firms and fixed contractual relations between them.
many regulatees decided not to co-operate, forcing the The emphasis in policy is also shifting from ‘control’
government to use other policy instruments. While to ‘generative policies’, which allow ‘individuals and
straightforward product rules might have been the groups to make things happen, rather than have things
13 Environmental Policy Instruments 263

happen to them, in the context of overall social con- cost criterion, between short-term costs (st) and
cerns and goals’ (Giddens 1994: 15). The value of long-term costs (lt). In multi-purpose instruments,
equality shifts from distributional equality, in terms of the environmental cost-effectiveness (or ‘eco-effi-
disposable income, to generative equality, in terms of ciency’) can only be established by attributing one
security, self-respect and self-realisation (Giddens part of the cost to environmental goals and other
1994: 191). These broader societal developments parts to each of the other objectives contributed to.
inevitably involve developments in environmental Other values relate to ethical categories of justice
policy instruments. and equality, covering traditional distributional jus-
What criteria can be used for the practical evaluation tice within and between generations, justice as fair-
of instruments? In assessing environmental policy, the ness, and the newer generative equality (on these
first criterion is probably environmental effectiveness. ethical issues, see Rawls 1972; Giddens 1994, 1998).
In the case of integrated environmental policies, how- Intergenerational justice has been made operational
ever, effectiveness cannot easily be established sepa- in an environmental context as part of ‘sustainabil-
rately from other values. How important is a toxic effect ity’ in the 1987 Brundtland Report. It is on a par with
on child development compared to cancer risks at a the discounting practice economists use, in which
later age and on a much longer time scale, and com- future effects count less than current ones.
pared to the loss of plant species that might have had However broadly the effects of environmental
pharmaceutical value? Hence, environmental effective- policy instruments are modelled, there will always
ness can be established only on the basis of broader, remain relevant aspects beyond modelling that have
non-environmental judgements. One such judgement to be taken into account and have to be specified as
relates to time. The time scale of effects requires choices second-order criteria or as strategic criteria (see
in terms of the relative importance of future effects. Table 13.1). The government has to operate with
The specific location of effects not only influences their some legitimacy, which means that, on average, some
type and magnitude, but involves different social groups minimum level of social and political acceptability
as well. The spatial distribution also relates to the way and support is required in instrument application.
effects abroad should be taken into account at home. Quantifying this aspect is difficult, but any environ-
Should national policies also aim at effects abroad as a mental policy that forgets about such aspects will
part of their overall effectiveness? WTO regulations falter in the long run. Furthermore, instruments have
run counter to such considerations, unless the rules to fit in to some extent with the capacities of the
apply equally to home products and imported products. existing administration. Emission taxes are easier to
For imported products not produced at home, environ- implement in countries with a tradition of effective
mental rules will mostly be seen as undue trade limitations. direct taxation. At a different level, major changes in
Another problem is how to deal with low-probability, sectoral competitiveness may create social instabil-
high-impact effects, where evaluating effectiveness is ity and should generally be avoided. Another ele-
based on the degree to which risk avoidance or precaution ment, which is lacking in most quantified models, is
is deemed important. how instruments influence technology development.
In addition to environmental effectiveness, itself Although these, partly overlapping, aspects are
value-based, there are also other values. One broadly hardly quantifiable, they may be essential for a well-
accepted value is that of costs, or rather, welfare founded judgement on environmental policy instru-
effects in terms of production losses required for ments, and for the long-term effectiveness of
environmental improvements. Instruments which environmental policy.
help stimulate environmental technology develop- Since instrument choice may bind society for years
ment, like market-oriented, pricing-based, economic or even decades to come, such choices are to be
instruments, will entail lower costs in the long run placed in a strategic context, not only taking account
(for the theoretical aspects, see Baumol and Oates of relations as they are now, but also of developments
1988, for existing instruments see Opschoor and Vos taking place in this longer time perspective. Four
1989 and for empirical aspects, see Hemmelskamp main strategic areas can be distinguished, relating to
1997). A clear distinction is thus to be made in the politics, social structure, economy and culture:
264 G. Huppes and U.E. Simonis

1. Instruments are to fit in with the broader conceptual 13.4 Design, Analysis and Evaluation
framework for public policy, e.g., along the lines of Policy Instruments
sketched by Giddens (1998).
2. They should be in line with the broader institutional
framework of society, e.g., in terms of increased Having described the role of environmental policy
mobility, functional specialisation of organisations instruments in a long-term perspective, and having indi-
and internationalisation of organisations. cated their role in policy and the criteria for their evalu-
3. They should take into consideration the general cul- ation, the final question is how to design, analyse and
tural developments, such as individualisation, mass evaluate them at a more operational level. It should be
culture and other-directedness, as outlined by many clear by now that there is no one unique way of doing
sociological studies. so. Many dimensions have their due place and not all of
4. Finally, instruments have to be adapted to general them can be included at the same level of aggregation.
economic developments, such as the globalisation of Four main dimensions can be distinguished, which
markets, shifts from hierarchical co-ordination to net- together create a framework for design and analysis.
work co-ordination, and shifts from the production of The framework may be used to generate a large number
commodities to the production of services, as outlined of distinct policy instruments, well over a thousand,
by Castells (1996). opening up options for well-argued choices on instru-
mentation. This section presents some existing examples
Some people, especially economists, simplify the evalu- of policy instruments, using the framework for their
ation by reducing it to an economic analysis. In principle, specification. The ultimate evaluation of designs and of
such an evaluation may cover all environmental effects; implemented instruments is based on their actual func-
it is based on a specific discount rate; it uses a risk tioning. Evaluation for the purpose of the design, or
avoidance factor of zero; it uses an equal weight for revision, of instruments has two main types of input (see
every euro or dollar, thus disregarding income distribu- Fig. 13.2 in the introduction to this chapter), one covering
tion; it takes only current private preferences into account; strategic points of view going beyond the level of the
and it assumes these preferences to be independent, case, the other referring to the analysis and evaluation of
meaning that nobody cares about the welfare of anybody the functioning of the instrument in its specific domain
else. In this situation, each emission or environmental of application, i.e. for particular cases. The discussion
intervention may indeed have an environmental price of these two types of input completes the chapter on
tag in terms of a (negative) net present value. Of course, instruments for environmental policy.
this also assumes that empirical consequences can be
fully specified in terms of items relevant to such hypo-
thetical individuals. Environmental policy instruments
can then be evaluated in one unit: money. This overall 13.4.1 A Framework for Design
score is the sum of the environmental effects, trans- and Analysis
formed into a net present value as outlined above, and
the direct economic (market-related) effects. The single Instruments as societal structuring mechanisms bring
euro or dollar figure resulting then indicates which order to the relations between actors and help to guide
instrument to use in which situation. their behaviour. What is common to all policy instru-
In reality, of course, this hypothetical situation does ments, and hence also to environmental instruments, is
not exist. Where price tags can be put on emissions, these that they have to bring about a change in behaviour relative
prices relate to partial effects and will mostly be based on to the behaviour without application of the instrument.
less than realistic assumptions. Several aspects of justice, In specifying instruments, we distinguish four main
such as equality and fairness, are omitted or included types of dimension. In question form, these are:
only superficially. Moreover, second-order and strategic
criteria are mostly not suitable for economic quantification. 1. Who influences whom?
Therefore, this option is too narrow to be the sole basis of 2. What is the influencing mechanism?
a convincing instrument evaluation, although costs of 3. What object is being influenced?
course do play a role in such an evaluation. 4. What is the operational goal?
13 Environmental Policy Instruments 265

These four empirical dimensions are quite general; in Although the general tasks of environmental policy
principle, they are the same all over the world, regardless instruments – avoiding the tragedy of the commons,
of cultural differences. In addition, and again in principle, solving the prisoner’s dilemma and preventing the free
these four dimensions can be analysed more or less inde- rider problem – could easily have been made into defin-
pendently, while other instrument dimensions seem more ing characteristics, they have not been included here
closely tied to specific cultures and institutions. Juridical because of their rather abstract and strategic nature.
status, for instance, is often used as a defining character- However, they do still play a role in instrument design
istic, but juridical categories are linked to the specificity and instrument evaluation, be it in general or at a case
of judicial systems. For instance, an EU regulation does level. The normative evaluation criteria specified in the
not have a counterpart in most other countries, while previous section have not been included here either, thus
Anglo-Saxon common law is not used in most European creating a distinction between empirical-descriptive ele-
or former communist countries. The systems converge ments of policy instruments and their normative evalua-
however, as new legislation in Anglo-Saxon countries is tion. Of course, there has to be a link between descriptive
statutory, while the courts in French law based European elements and evaluation, since it is ultimately the evalu-
countries have increasing interpretational power. ation that counts. The evaluation is based not only on
Using these four dimensions, can we now specify in the direct effects of the instruments; decisions should be
more detail what makes instruments into environmental guided by the ‘ultimate’ effects. There is a tendency to
policy instruments? The relations between actors, i.e., include some standardised effect mechanisms into pol-
who is influencing whom, might give some clue. One icy instruments, like including global warming potential
could argue that instruments issued or used by a minis- (GWP) in national and international climate policy.
try or department of the environment are environmental Thus, some mechanism may play a role in the goal spec-
policy instruments, using the first dimension. This ified in the instrument. Most social and environmental
would mean that an administrative reorganisation, for mechanisms, however, will be independent of the instru-
instance shifting policy-making in this area to the depart- ment, which means that their analysis must be included
ment of agriculture, could change their status. It would as a separate step in the evaluation of policy instruments,
also mean that social instruments could never be envi- not in their definition.
ronmental instruments. The second dimension, the The four main dimensions chosen now need to be
influencing mechanism, is not suitable to define the defined in greater detail. It should be clear that there is
environmental nature of a policy instrument either. not one general truth at this level either. For instance,
Permits, taxes and excises and prohibitions have a general ‘actors’ can be described in many dimensions, for
nature in all regulatory contexts. The third dimension, instance as individuals or firms, while firms can be
the object influenced has no specific environmental sta- described as small or medium sized firms, large national
tus either, as there are many rules on products and instal- firms or multi-nationals, etc. What would be the guiding
lations not related to the environment. It is the fourth principle for such further choices? In the end, the ques-
dimension, the operational goal, indicating the direction tion that has to be answered is how policy instruments
of behavioural adjustment, that decides whether an can fulfil their function of simplifying the complexities
instrument is an environmental instrument: its goal has of reality to allow effective and concerted actions
to be an environmental one, or at least its ultimate aim towards environmental goals, and distinctions should
should be. This is not always a clear-cut criterion. If, for serve this ultimate purpose. However, the purpose of
example, a government stimulates hydropower with the this section is also very practical, namely to decide how
intention of becoming less dependent on imported fossil real instruments can be created and how their expected
fuels, this also helps to reduce CO2 emissions. Hence, functioning can be evaluated. In approaching this task,
the operational goal of stimulating hydropower need not there is a tendency to introduce further distinctions rel-
be based on environmental considerations. Such border- evant to the situation. As our four main dimensions
line cases will certainly exist in practice. Even if none of already lead to over a thousand instrument categories,
the participants sees a particular instrument as an envi- further systematic detailing, however relevant it may be,
ronmental policy instrument, it could still be categorised should be used sparsely if at all. Such relevant additional
and analysed as such, relating the instrument to this goal aspects are more suitable for the evaluation of, in this
and ultimate aim. sense, more sparsely defined instruments.
266 G. Huppes and U.E. Simonis

[Link] Who Influences Whom? [Link] What Is the Influencing Mechanism?

In deciding who is influencing whom, a major distinc- The influencing mechanism specifies how one actor
tion can be made between governments on the one hand influences the other. We follow the main categories
and non-governmental private actors, like individuals, distinguished by Bressers and Klok as explained in the
firms and organisations, on the other. These two types of previous section, with some further differentiation as
actors involved in instrument application lead to a further found in regulatory practice, see Table 13.3 for a sur-
categorisation of instruments, involving a distinction vey. The first mechanism involves changing the set of
between three types of actor relations (see Table 13.2). available alternatives, for instance by their prohibition
This section uses the three main types of actor or prescription, or by the creation of new options.
relations to categorise the instruments as political- These define the first three instrument categories.
administrative instruments, regulatory instruments Secondly, the consequences of alternatives can be
and social instruments. changed, for instance by influencing market prices and
Regulatory instruments are the most common type. market volumes. This is the category of economic
An environmental permit is a major instrument for gov- instruments. The third major mechanism is the change
ernments to influence private actors (including publicly in the evaluation of consequences, achieved for
owned firms), as are emission taxes, such as SO2 taxes instance through information on the effects of actions
creating a market incentive for reducing SO2 emissions. and through normative guidance. These are the cultural
Political-administrative instruments may work at an instruments. In addition to these actor theory based
international or national level. An international treaty categories, we distinguish the influence exerted by
like the Montreal Protocol on substances that deplete changes in the institutional structure of society, as
the ozone layer is an instrument between governments, structural instruments influencing actions. Liability
while an EU environmental regulation binds national rules are a major example. Finally, there are the proce-
governments in the EU. The ISO 14001 industry stan- dural influences, functioning as procedural instru-
dard is an example of a social instrument. By specify- ments. They are introduced separately here, as their
ing rules, it guarantees that environmental audits have a use is widespread in specifically environmental con-
degree of generality and reliability in describing the texts, as in covenants and audit systems.
environmental performance of firms, creating an incen- The influencing mechanism is to be clearly distin-
tive to take environmental aspects seriously. Other guished from mechanisms further down the causal
examples are private certification systems, such as chains. Such further effects may well involve the same
those used for food in supermarkets, influencing the
behaviour of food producers and creating options for Table 13.3 A typology of instrument mechanisms
environmentally oriented choices by consumers. Mechanism-based
instrument specifications Examples
Prohibiting instruments • No cadmium stabiliser allowed in
Table 13.2 Actor relations based instrument specifications
PVC as a building material
Name of Prescriptive instruments • Legal obligation for separate
Actor relations instrument Examples waste collection
Governments influencing Political- • Montreal Option-creating • Multiple waste containers for
governments administrative Protocol instruments separate collection
instruments • EU regulations Economic instruments • Volume: auctioned car ownership
rights
Governments influencing Regulatory • Environmental • Price: energy tax, SO2 tax
private actors instruments permits Cultural instruments • Normative: ecolabel on products
• SO2 emission • Relate organic solvents to
charges summer smog
Private actors influencing Social • ISO 14000 Structural instruments • Liability rules
private actors instruments Series • Public decision-making safeguards
• Private Procedural instruments • Obligatory environmental officer
certification in firm
systems • ISO 14001 audit
13 Environmental Policy Instruments 267

mechanisms, but do not form the basis for instrument products

categorisation. For instance, a change in liability rules
will ultimately involve market mechanisms and the single object installations
creation of options with lower liability risks. The influ-
encing mechanism, however, is that of a structural activities
instrument. The terminology used in practice varies objects
somewhat. For instance, the prohibition and prescrip- regional
products
tion of options is also referred to as direct instruments global
or as juridical or legal instruments. This terminology classes of regional
installations
seems awkward, however, as option creation can be objects global
seen as a direct instrument as well, and financial instru- regional
activities
ments like emission taxes also have a distinct legal status. global
The combination of prohibiting and prescriptive instru-
Fig. 13.4 A classification scheme of regulated objects
ments is also referred to as binding instruments.

Table 13.4 A typology of influenced objects

[Link] What Object Is Being Influenced? Object influenced Examples
Single product • Single aircraft flying permit
A further dimension is the nature of the object being Single installation/firm • Testing whether a building
affected by the influencing mechanism. A first basic harmonises with the landscape
distinction in objects being regulated is that between • Permit requiring safety valve on
some pressure vessel
single objects and classes of objects. This relates to the Product classes, regional • EU rule on compulsory three-way
applicability of the instrument and to the options for catalytic converters in cars
implementation. A second distinction is that between • Rules on NOx concentrations from
products, which are mostly mobile physical entities; household boilers
installations, which are mostly immobile physical enti- Product classes, global • WTO rules on non-discrimination
Installation classes, • German rules on allowable SOx
ties; and activities, focussing on more general behav-
regional emissions per kWh at power
ioural options. Such behavioural options will usually generation
be linked to some physical object, like speed limits for Installation classes, • IAEA rules on safety requirements
cars in a city. These objects may be defined in very global for nuclear installations
abstract terms, however, like environmental auditing Classes of activities, • Local speed limits for passenger
regional cars
rules for firms, relating to all their installations. In the
Classes of activities, • ISO 14001 requirements on
US, most regulated physical objects are called installa- global environmental planning in firms
tions. Since no example of a single activity being regu-
lated comes to mind, this class has been omitted. As
the classes of objects are necessarily restricted as to as such. However, regulating things may be easier than
spatial scale level, a further distinction is made based regulating activities, as most things can be inspected at
on classes of objects being regional or global. This any time, while effective inspection of behavioural
results in eight types of object (see Fig. 13.4 and Table aspects is much more complicated. Measuring the con-
13.4). While no specific instrument names are linked centration of NOx in a boiler outlet requires real-time
to these classes, the term product policy is becoming measurements, so it is easier to prescribe a burner type.
popular in EU policy and in a number of countries, Cadmium in PVC stabilisers in building materials,
referring to classes of products in the EU. however, can be measured at any time. For the sake of
Regulating material objects, as ‘things’, is not done regulatory efficiency, requirements are often specified
because of their inherent properties. Ultimately, it is in terms of the technical composition of the product or
only the processes in which they function, as activities, installation, assuming that these will result in an emis-
which influence the environment by causing environ- sion reduction. In a considerable proportion of cases,
mental interventions. Environmental policy instruments however, this relation is not fixed. Bypassing a flue gas
try to influence these interventions, not the techniques purification installation, for instance, saves costs. Illegal
268 G. Huppes and U.E. Simonis

bypassing can be stopped only if the inspector calls at across various policy themes, has been occurring over the
the right time. last decade, like the use of some eco-indicator as an
Classes of activities may be defined narrowly, as with integration of similar effect types, or even integrating all
denied access to National Parks in Germany after sunset, environmental effects in a normatively based overall
or broadly, as in the environmental auditing of all firms environmental score. An overall evaluation of all envi-
in the world. A good deal of complexity can be intro- ronmental interventions makes the relative importance
duced in defining classes of activities. LCA (Life Cycle of various environmental effect categories explicit. By
Assessment) refers to all processes (= activities) involved allowing a shift between them, based on the explicit
in having the function of a product delivered. In environ- trade-off, it is their combined effect which is regulated.
mental audits of firms, it is becoming common to incor- The NOGEPA (Netherlands Oil and Gas Producers
porate the supply chains involved. This is similar to Association) covenant in the Netherlands specifies such
LCA, as the object is also a system of linked activities. an integration. It forms the basis for one further integra-
Such larger entities are attractive, as a policy instrument tion step, combining environmental and economic goals,
may influence all of them in a balanced way. Since some in order to select the most eco-efficient measures for
arbitrariness is involved in defining such systems, bind- environmental improvement.
ing measures may be more difficult to apply. The closer the instrument goals are to environmental
aims, the more efficient the instrument can be, by avoid-
ing unnecessary technology fixation. If a policy instru-
[Link] What Is the Operational Goal? ment uses the mass of a car as a proxy for its emissions,

Operational goals specify what the instrument states as Table 13.5 Typology of operational environmental goals in
the objective to be approached or achieved. ‘Cadmium instruments
in PVC window frames’ may be forbidden, as in the US, Operational goals Examples
or a ‘weighted maximum of emissions’ may be allowed, Composition of product or • Cadmium stabiliser in PVC as
as in a number of Dutch environmental covenants. We installation a building material
can categorise the goals in relation to how close they are to • Percentage of post-consumer
ultimate environmental aims, see Table 13.5 for a survey. paper waste in paper
Most traditional binding environmental policy instru- Technology characteristic • Double skin in oil tankers
• Inherent safety in nuclear
ments, like the operating permit, have relied heavily on installations
easily verifiable rules for the composition of a product • Take-back legislation (several
or installation, like banning the application of PCBs, or European countries)
prescribing a filter on air outlets, or banning some mer- Single environmental • Noise-based landing fees in
cury-based technology for chlorine production. Since intervention airportsCO2 tax
• Ambient air quality standards
the 1980s, the emphasis on efficiency and problem pre-
Set of environmental • Set of allowable emissions as a
vention has induced a shift towards goal specification interventions goal permit
closer to the ultimate environmental goals. Goal-oriented Single theme score • Emission reduction targets
permits specify the maximum emissions of the installa- (Kyoto Protocol)
tion, or the maximum emissions per unit of product pro- • Environmental policy plans
duced, as in SOx per kWh of electricity produced in Set of theme scores • LCA-based rules for waste
prevention (Germany, Netherlands)
Germany. Policy integration, as a central element in effi- Effect-oriented eco-indicator • Limited set of eco-indicators in
ciency, has led to further steps, through integration in scores building regulations
terms of policy themes like global warming and acidifi- Single integrated • Best Available Technology
cation. The Kyoto Protocol includes steps to integrate environmental score specification (EU)
various substances in terms of their contribution to the • EPS based design rules in industry
• Theme-based weighting
policy theme of climate change, based on their Global
(NOGEPA covenant)
Warming Potential. Estimates by the World Bank indi- Combined environmental • Eco-efficiency as a selection
cate that such integrated policies, as opposed to the and economic score criterion for emission-reducing
focus on CO2 emission by energy systems, can reduce technical measures (NOGEPA
the cost of climate policy by 60%. Further integration, covenant)
13 Environmental Policy Instruments 269

the regulation does not influence the type of engine able, like using a panel of officials and experts (e.g.,
characteristics or the driving style. If only the mass NOGEPA covenant, see Huppes et al. 1997); using
indicator is influenced, cars will get lighter but not nec- policy aims for weighing emissions into one score
essarily cleaner at the same rate. If emissions are taxed (e.g., Swiss or Norwegian Ecopoints); using a mixture
or bound to a maximum, these other characteristics will of partial economic valuations or some equivalency
primarily be influenced, with probably a limited effect factors (e.g. ExternE and EPS); and applying some
on car mass. Measurement techniques for such instru- preferences or value types, see Guinée et al. (2002) for
ments are not yet available but may well be developed a survey.
in the near future. All these practical models have been developed in
As regards integrated policy goals in instruments, limited domains, in hardly peer-reviewed studies, with-
there is a gap between what modelling can more or less out a broad public discussion. Important questions are
realistically achieve (with adequate validity and reli- only touched upon and not answered. How can we dif-
ability), and what is needed for integration in the single ferentiate between reversible effects, like ecosystem
environmental score. Somehow, the modelled multi- degradation, and irreversible effects, like species extinc-
tude of environmental interventions and other effects tion? How can we differentiate between low-probabil-
have to be transformed into an overall judgement. ity-high-risk effects, such as possible runaway effects
There are several methods for this purpose, which do in climate change, and more probable slow-change sce-
the undoable. Economists derive overall evaluations in narios? Assuming that uncertainties can be specified in
monetary terms on the basis of past behaviour and the terms of risk or at least subjective probability, how can
stated preferences of individuals. In a very different we evaluate options with different probabilities? Even
approach, impact assessment in LCA first integrates, in the extremely well researched area of climate
on the basis of modelling, into several environmental change, surprises and outright disasters have not yet
policy themes and then through a weighting procedure been incorporated in policy models, though the first
merges these into an overall evaluation. In whatever frameworks are being developed (see Schneider and
way this integration is achieved, for reasons of policy Kuntz-Duriseti 2002). Given the uncertainties involved
consistency it would be necessary to use the same trick for each policy theme, how can we make a compara-
every time. There is a modest requirement for the over- tive evaluation of climate change effects, which can
all rationality of environmental policy, which is that hardly be specified in economic (i.e., welfare) terms,
the trade-offs between various effects are equal for as against effects of acidification resulting in reduced
each policy. This simple requirement can be trans- crop yields and increased corrosion, which can quite
formed into a conditional statement (see von Neumann easily be specified in terms of economic losses? Such
and Morgenstern 1953; Sen 1982): if policies are ratio- fundamental problems have yet to be solved before
nal, there is a single set of weights on their effects integrated goals can be used more widely in policy
which can ‘explain’ all policy choices made. If this instruments and in policy.
condition is not met, this means that it would have
been possible to reach the policy aims in a less costly,
that is, more eco-efficient way. Given the theoretical [Link] Combining Actor Relations,
and practical limitations of modelling, policy integra- Mechanisms, Objects and Goals
tion can only be reached through practical choices,
based on incompletely developed arguments. Making Combining the four main dimensions, those of actor
a start here is better than accepting the even less attrac- relations (3); influencing mechanisms (7); objects influ-
tive current state of affairs. enced (8); and goals (9), results in a large number of
When making this start, the best one can do is to instrument types: 3 × 7 × 8 × 9 = 1,512. Not all combi-
strive for a consistent and transparent solution, on the nations may be relevant, however, or even possible. A
one hand taking into account the as yet only partly technology specification, as a goal, cannot easily be
known real mechanisms, and on the other hand speci- applied to ‘classes of activities’, as an instrument object.
fying the normative background of the evaluation. This The combination of more binding instrument mecha-
problem area, it seems, has not yet been under serious nisms with more encompassing objects of regulation
scientific scrutiny. Some practical solutions are avail- and with more integrated policy goals makes policy
270 G. Huppes and U.E. Simonis

development and implementation increasingly com- policy instrument creating the incentive to use it.
plex. However, the problems involved in this integra- Mostly, such tools are incorporated in social instru-
tion at the instrument level are mainly the same as those ments or in weak regulatory instruments. Thus, the
at a policy level using several instruments. Criteria like potentially most efficient instruments can only play a
consistency are very visible at the instrument level and very limited role at present, while the not-so-efficient
not at the policy level. The aim would be to have con- regulation of technologies by prohibiting and prescrip-
sistency at the latter level as well, to avoid unnecessary tive instruments still carries the main burden of
costs in achieving environmental aims. environmental policy implementation.
The aim of consistency in environmental policy, and When specifying integrated goals for complex
the related aim of eco-efficiency, may be achieved objects in policy instruments, it should be clear that
through adapted markets, as by emission taxes, tradable instruments cannot comprise the full extent of all real
emission permit systems and liability rules. Assuming effect mechanisms. The creation of direct effects inevi-
some degree of competitiveness in the markets, the nor- tably leads to a whole range of direct and secondary
mative information on policy aims and the empirical effects on society and the environment, not all of which
relations in markets and technologies are combined in can be incorporated. It is not even the most sophisti-
the price structure faced by individual decision-makers cated models available which play a role in instruments,
in society, and taken into account in their decisions. as in covenants in Germany and the Netherlands. The
Such broadly applicable ‘macro-instruments’ (see Huppes simplified standardised modelling in these instruments
et al. 1992; Huppes 1993) may, however, not be ready is an approximation, which should not be confused
for widespread application, for administrative-technical with state of the art modelling of real effects. By being
as well as political reasons. The market-based informa- operational at a decentralised level, the approximation
tion system, as Adam Smith’s invisible hand, is there- may represent an improvement on other ways of regu-
fore mainly lacking, guiding economic decisions in the lating technologies and products, which may not take
wrong directions from an environmental point of view. the simple effect mechanisms into account at all.
It is to be created by environmental policy, as efficiency Fortunately, things in real world situations are
is a generally accepted central goal. All the more com- sometimes less complicated than they are theoretically
plex object types and goals types have been set up to perceived to be. In such cases, environmental policy
create this integrated view on environmental regulation. instruments can be simpler as well. For example, when
Attempts to avoid problem shifting to other emissions, banning a toxic and persistent agrochemical for which
to other policy themes and to other times, which are slightly more expensive alternatives are available, the
preconditions for efficiency, constitute a key element in real effect route in the economy hardly has primary or
a tool like LCA, which is beginning to be incorporated in secondary effects in the chain, nor is there much com-
policy instruments. plexity in the environmental pathways towards valued
In using the more integrated classes of activities as endpoints in terms of human and eco-toxic effects.
objects, their definition as a system with internal rela- There is then no reason to complicate the instrument
tions has become a research subject in itself, that of and burden it with complex effect mechanisms and
environmental systems analysis. Major tools for this evaluations. A simple prohibition of the agrochemical
purpose are SFA (substance flow analysis), LCA (life will do, after a relatively simple analysis of effect
cycle assessment) and E-IOA (environmentally chains in the policy formation process, including a
extended input-output analysis). They are mainly sim- check on the availability of not too costly alternatives.
ple models, based on linear relations and constant Such easy solutions have mostly already been imple-
technological relations. However, the fact that they are mented, however. After more than 30 years of active
simple, with a limited validity and reliability, means environmental policy, it seems that most simple
they can be made operational. Where decisions on end-of-pipe (add-on) measures and simple product
technologies and markets are made at a decentralised prohibitions have already been enacted. Such policies
level without much direct government influence, as is may now even start to hamper environmental progress
increasingly the case in a globalising world, we need a by fixing old technologies. The remaining problems
decentralised tool that comprehensively indicates the are more complicated and may well require a more
ultimate effects of such decisions, incorporated in a sophisticated instrumentation.
13 Environmental Policy Instruments 271

in pricing. The result is the ‘noise related airport landing

13.4.2 A Thousand Instruments Defined,
fees’, which is a fairly common social instrument used
with Examples by airports near larger cities. The motivation behind a social
instrument may be another policy instrument, especially
A policy instrument may be defined by combining a regulatory one. In the example given here, the motiva-
elements from the four basic instrument dimensions tion-creating regulatory instrument may be the operating
discussed above, in principle leading to well over a thou- permit of the airport, stating maximum noise levels in
sand major instrument types. These dimensions are more surrounding residential areas, which is a prescriptive
or less independent, so they may be used as a framework regulatory instrument for a single installation, involving
for instrument development, as an instrument generator. both technology characteristics and sets of environmen-
Any combination defines the main lines of an instrument. tal interventions as its goals. A second motivation step,
Figure 13.5, ‘the instrument generator’, gives some exam- behind this regulatory instrument, may be a political-
ples. Take, for instance, ‘social instrument’ from the actor administrative instrument, such as an EU directive on per-
relations column; use ‘economic instruments’, used here missible noise levels in residential areas. Another
in the sense of pricing, from the set of instrument mecha- example, from the US, is the extended liability which has
nisms; take ‘product classes, global’ from the set of objects been achieved for toxic wastes that have not been treated
influenced; and take ‘single environmental intervention’ properly, to avoid damages to the environment. It involves
from the set of operational goals. This instrument then a structural mechanism and pertains to all waste sites
can be further specified in terms of the product classes, with toxic wastes in the US. All highly hazardous
for instance using ‘aircraft’, with different noise levels, substances are involved, as potential environmental
measured in a specified way, as the fully operational goal interventions. Another example is the Kyoto protocol,

Operational goals:

composition of
Object influenced: product or
installation
Mechanism based
instruments: single
product technology
characteristic
prohibiting
instruments single
installation single
environmental
Actor relations: prescriptive intervention
instruments product classes,
political- regional
set of
administrative option-creating environmental
instruments instruments product classes,
global interventions
regulatory economic single
instruments instruments installation classes,
regional theme score
social cultural set of
instruments instruments installation classes,
global theme scores
structural effect-oriented eco-
instruments classes of activities,
regional indicator scores
procedural single integrated
instruments classes of activities,
global environmental
score
Legend:
maximum climate-changing emissions per country (Kyoto combined
Protocol) environmental and
noise component in airport landing fees (Amsterdam Airport) economic score

extended liability for final toxic wastes (Superfund US)

Fig. 13.5 The instrument generator

272 G. Huppes and U.E. Simonis

which limits the total amount of climate changing emis- These options seem to stretch the concept a bit.
sions in terms of their global warming potential. These However, their basic function is to allow states to
limits are set for countries, involving all emitting activities develop regulatory activities which would not be pos-
in each country, as one level of geographic region. sible or at least rather unlikely without the explicit
Using only the first two dimensions, for a start, results in development of the option.
27 instrument categories. An example of each is given below.
3. Economic instruments
I Political-administrative instruments (guiding
• Internationally tradable emission reduction obli-
relations between public bodies)
gations for climate changing emissions, as might
The meaning of the word government is restricted here be based on the Kyoto protocol
to the regulatory part of governments, engaged in plan-
4. Cultural instruments
ning, developing and implementing policies, and using
policy instruments in such policies (in this case environ- • International guidelines, as by OECD, and in the
mental policies). Other operative public tasks, like build- EU IPPC/BAT (Integrated Pollution Prevention
ing and maintaining roads, canals and dikes, maintaining and Control/Best Available Technology) rules
an army and distributing electricity, are productive or • Rio Declaration, AGENDA 21
consumptive activities, to be regulated like any other • National indicative guidelines for local zoning
economic activity. The relations between governments laws
as regulators, and private persons and organisations as • ILO conventions regarding environment-related
regulatees, are in most cases hierarchical. The relations labour standards
between governments may be hierarchical as well, as
5. Structural instruments
when the EU binds the policies of countries with
Directives, and national governments prescribe policies • WTO rules on environmental considerations in
to regional and local governments. However, in the inter- restrictions on trade
national context, most relations between environmental
6 Procedural instruments
policies are horizontal, as in bilateral and multinational
treaties. Some hierarchy is implied when international • International Criminal Court (ICC) (no environ-
public bodies are involved. In addition, where bilateral mental example available)
relations seem to be involved, there may still be a hierar-
chy. For instance, while ‘joint implementation’ is dealt II Regulatory instruments (guiding public regulator
with at the interstate level, the rules for joint implementa- – private regulatee relations)
tion are dealt with in the Kyoto Protocol, and the possible
future extensions to that protocol might be designed hier- Regulatory instruments guide regulator – regulatee
archically under UN leadership. relations; they are the traditional environmental policy
The prohibiting and prescriptive instruments are instruments.
combined here as Binding instruments. In principle, 1. Prohibiting and prescribing (= binding) instruments
the seven main implementation mechanisms discussed
• Binding instruments can be either prohibiting or
above can be applied.
prescriptive. Prohibiting instruments are usually
1. Binding instruments conditional, in that something is forbidden
• International treaties and conventions including unless some requirements are fulfilled. In sym-
binding elements, like the Montreal Protocol, the metrical situations, as with speed limits, the dif-
Kyoto Protocol and the Biosafety Protocol ference between prohibiting and prescribing
• EU environmental directives for member states instruments is small: it is forbidden to drive
faster than the limit or prescribed to drive more
2. Option-creating instruments slowly.
• Clean development mechanism, as under the • Allowable coolants in household refrigerators
Kyoto Protocol • Operating permits for installations, the work-
• Multilateral Ozone Fund under the Montreal horse of environmental policy
Protocol • Land use regulations and zoning laws
13 Environmental Policy Instruments 273

2. Option-creating instruments 1. Binding instruments

• Separate waste collection facilities, in all Western • Contractually specified rules for waste man-
countries agement, as when firms commit themselves to
delivering a certain amount of waste over a longer
Option creation can be direct, as in providing separate
period of time (Netherlands)
waste collection facilities to households that may vol-
untarily separate their wastes. No prescription or pro- 2. Option-creating instruments
hibition is involved.
• Battery take-back facilities in supermarkets
3. Economic instruments: price-changing (financial)
3. Market instruments
instruments
• Noise-related landing fees at airports
• Emission taxes (or: charges, levies, excises) on
• Deposit-refund system on cadmium-containing
CO2 (e.g. Norway) or on SO2, formerly in Japan,
rechargeable batteries for household appli-
now in China
ances, on a voluntary basis
• Road pricing (most Western countries)
• In-firm tradable emission permits (see Box
4. Economic instruments: market volume instruments 13.4)
(‘things’ only)
4. Cultural instruments
• Tradable emission permits, like SOx permits in
• Green marketing (all Western countries)
the US
• Green accounting (ISO)
• Tradable production rights, like fishery rights
• Ecolabelling rules (ISO)
• Tradable product ownership permits, like car
permits in Singapore 5. Structural instruments
5. Cultural instruments: non-compulsory structured • Standard contracts specifying adherence to
information environmental standards, set up for instance by
a branche organisation
• Public eco-labelling schemes
• Public certification of firms, as for refrigerator 6. Procedural instruments
repair firms in the Netherlands
• ISO 9000 Series, on quality control
6. Structural/institutional instruments • ISO 14000 Series, on environmental perfor-
mance measures
• Extended liability
• Good housekeeping ownership rules The international standard on environmental auditing,
• Educational system, Copernicus charter, etc. ISO 14001, for instance, is a procedural instrument,
requiring firms to take due notice of environmental
7. Procedural instruments
aspects in their operations, in the sense of having an
• Covenants, voluntary agreements ‘in the shadow environmental policy plan, having officials responsible
of the law’ for checking its progress, etc. If rules were incorporated
• Environmental Impact Assessment rules on how to further specify environmental performance,
• Obligatory information disclosure, as in the US the instrument would become a cultural instrument.
Toxic Releases Inventory (TRI) Act
III Social instruments (guiding relations between
private actors)
13.4.3 Choice of Instruments in Policy
These instruments are similar to political-administra- Design
tive instruments in that they may reflect horizontal
relations between equals, or have a hierarchical ele-
Using the framework for environmental policy instru-
ment in them, as is often the case in environmental
mentation and the evaluation criteria indicated above
supply chain management. Again, the six main imple-
would seem to suffice for a rational development of
mentation mechanisms apply.
274 G. Huppes and U.E. Simonis

Box 13.4 Multinational company tradable emis- where a company has competitors with less strin-
sion permit gent policies, its costs will show a relative rise.
Some major oil companies, including Shell, have Hence, in the course of time, there will be a shift to
introduced emission trading between the firms com- firms not participating in this (or a similar) emission
prising these multinational companies. The emission permit trading scheme. Also, questions arise as to
trading focuses on climate-changing emissions, like how company environmental policy relates to public
carbon dioxide and methane. Each independent busi- environmental policies in the various countries
ness unit within Shell has an number of emissions where the company operates. If more stringent poli-
rights, which may be sold to other Shell units. There cies are introduced in some countries, the permit
is an accounting system which establishes the actual system will no longer have effects there, as induced
emissions of each unit. Emissions without a permit costs of emission reduction are higher than the per-
are not allowed, resulting in a company-internal cost mit costs. In countries with emissions taxes, the
penalty. If a unit has more permits than it needs, it will firms involved will have a greater incentive to reduce
try to sell them to other units. If it wants to expand, it emissions than other firms in the company, reducing
may acquire permits on the internal but global com- the overall efficiency within the company. In this
pany market. The total amount of emissions permit- sense, companies using such a scheme will create an
ted is being reduced slowly, based on the company’s argument against more stringent national policies.
environmental plans, by reducing the allowable emis- Conversely, if public policies are less stringent than
sion volume per permit each year. The combination the company scheme, they become superfluous. In
of business expansion and the reduction of the overall this situation, multinationals like Shell create an
emission volume permitted results in upward pres- incentive for national governments to implement
sure on permit prices, while environmentally oriented more stringent policies. The overall effect will be
technological development leads to a downward pres- that public policies will tend to be harmonised at a
sure. What will the effects of this instrument be? global level towards the level of emission reduction
The effects in terms of company emissions are indicated by the large multinationals. If most multi-
quite clear: the goals of Shell’s environmental policy nationals were to come up with similar and equally
plan are met, while leaving the business units their stringent schemes, there would be a clear drive
technological freedom. The emission reduction is towards uniform policies, at the level of stringency
achieved in the most efficient way, as each business chosen by those firms, and not by governments. It
unit reduces its emissions to the level where cost should be relatively easy to extend the tradable per-
reductions are (roughly) equal to the costs of having mit system to trade between firms. The choice of
the permit. A major problem in implementing such their policy instrument will influence policy imple-
a system is choosing the system boundaries. How mentation by governments as well, making it very
can firms partly owned by Shell and partly by other difficult for instance to implement emission taxing
companies participate in the scheme? What happens schemes on top of the company tradable emission
to the total number of Shell permits if Shell sells permit scheme. Shell chairman Moody-Stuart has
some of its activities, or acquires others? called upon governments to implement similar mar-
What the net environmental effects in the global ket-based mechanisms to achieve their Kyoto tar-
society will be, in terms of reduced climate-chang- gets. For further information on the Shell tradable
ing emissions, is less clear, which in this case is due emission permit system (STEPS), see [Link].
to quite complex indirect effects. For activities com/climate.

policy instruments for the environmental problems and guide the process, possibly leading to different
facing us. The question may be asked, however, how results. In this final part of the chapter we comment on
the analytical approach depicted here relates to process two such preliminary considerations, those underlying
aspects in instrument and policy design at a case level. the definition of the case, and some case-independent
This also involves preliminary choices that may start considerations guiding instrument choice (Box 13.5).
13 Environmental Policy Instruments 275

Box 13.5 Main lines of argumentation in Section effectiveness of such policy. Focussing on social
13.4 procedures in regulatory instruments may enhance
• Policy instruments are not given entities to be the effectiveness of specific policies, but implicitly
examined; they are social constructions with excludes other policies from being developed and
many degrees of freedom. implemented.
• Four main dimensions are central to the definition • Structural instruments like liability rules and
of specific instruments, though probably not emission taxes may exert their influence at low
enough for a full specification of operational transaction costs and with potentially high envi-
instruments. They are: the nature of actor rela- ronmental effectiveness, but for the time being
tions; the instrument mechanism in implementa- only on a limited number of environmental
tion; the objects influenced; and the operational effects.
environmental goals embodied in the • In design and evaluation of policy instruments
instruments. and policies, one part of the analysis is empirical
• After specification, these four dimensions create while the other is normative.
an ‘instrument space’. Criteria, ultimately evalua- • The empirical analysis is partially subjective and
tion criteria, signpost the route through this instru- concerns direct, indirect and, as far as possible,
ment design space for relevant instrument secondary effects. No broadly accepted models
choices. are available, leaving much room for complex
• Instruments are building blocks in the process of debate.
policy formulation and policy implementation; • The criteria for instrument and policy evalua-
they are not the policy itself. tion refer to direct expected effects, but also
• In actual policies, public and private, consensual include second-order criteria and strategic crite-
acts are at the core of behavioural adjustments. ria, viewed from a long-term perspective of the
This should not obscure the fact that power and development of environmental policy
interests play key roles in such processes and instrumentation.
that power is very much based on the availability • Basic choices on the desired nature of regulatory
of operational policy instruments. instruments, in view of overall institutional devel-
• Given some maximum regulatory effort for envi- opments in global society, would have a direct
ronmental policy, there is a limit to the overall bearing on instrument choices in specific cases.

[Link] Cases the traditional operating permit in mind, one would

probably define the problem this way. By contrast, if
Environmental policy instruments can best be chosen the problem is defined in terms of a lack of oligotrophic
and designed in greater detail by specifying the char- areas, or as a lack of differentiation in nutrient concen-
acteristics most relevant to the case at hand. However, trations, a regional regulatory scope will be more logi-
it is not always easy to say what exactly is ‘the case’, cal. If one already had a different instrument option in
as the case is defined both by the problem context and mind, like a substance deposit (see Huppes 1988), the
by preliminary choices made before a specific instru- scale of the problem could well be defined at a regional
ment is designed. These reflexive or interactive rela- level. The use of individual permits then no longer
tions make instrument design a much less rational and seems so obvious. Evidently, it is not only the empirical
mechanical activity than might seem possible at first context that indicates choices; normative considerations
sight. If, for instance, one first defines the ecological on how to regulate may well already play a role in the
effects of eutrophication as a manure problem, and problem or goals definition phase. The polluter-pays
then the manure problem as a consequence of too many principle reflects the normative principle that he who
animals per hectare, the choice to regulate the number pollutes has to pay for the consequences of his action
of animals per farm seems logical. If one already had – and for the costs of preventing them, as is the case in
276 G. Huppes and U.E. Simonis

liability law in all Western countries. Tradable permits social context in which they are functioning. A major
lead to prevention costs being borne by the polluter, difference, for instance, is that between the litigative
but not the damages. In this respect, an emission tax is American style of regulation, where laws are often
more in line with general social and legal consider- fiercely debated and enacted after lengthy litigative
ations than a tradable permit. If it is specified as a procedures, and the more horizontal policy process in
regional substance deposit, the focus on individual England and the Netherlands, with officials influenc-
emitters vanishes more or less completely. With such ing private decisions through discussion and informa-
caveats in mind, we now turn to design choices in tion, and only ultimately through threats of harsh
instrumentation, within the framework developed. regulatory actions against non-cooperative regulatees
(cf. Jänicke and Jorgens 1998; Vogel 1986). In such
systems, implementation may take place ‘in the shadow
[Link] Process Aspects of Design of the law’, without any new laws or permits being
enacted, or in a private context, with contracts or cov-
A possible objection to the above procedure might be enants being signed.
that in this top-down approach, the uniqueness of In administrative science, this has led to greater
actual problems and options for their solution will be emphasis on the process of policy formulation and
lost. This is a valid objection. The answer is that for implementation, reducing the emphasis on the more
any institutionalised policies, the knowledge of the formal characteristics of policies in terms of instru-
concrete cases is so far removed from the regulators mentation. Policy-making is then easily regarded as a
that real knowledge of ‘the case’ will usually not be discursive process between all those involved, with
available. Another objection might be that, as there are outcomes in terms of their environmental actions based
so many relevant evaluation criteria, these at least on the power, interests, resources and the shrewdness
should be reflected in the design space developed here. of the actors involved. The present authors prefer what
Such further criteria may indeed help to specify actual they regard as a balanced view in this respect, indicat-
instrument design in a relevant way. Another objection ing the role of policy instruments both in terms of
may be that the top-down approach is a hierarchical, structuring discussions and as indispensable means to
technocratic approach, even an anti-democratic one. wield power and shape both society and the environ-
Analytic hierarchy, however, is not necessarily linked ment. Of course, this does not deny the fact that poli-
to social top-down approaches. On the contrary, the tics play an essential part in policy development, nor
quality of the democratic process may well be improved that social processes are fundamental in terms of both
by reducing the complexity of the subject of regulation policy-making, including instrumentation, and policy
while at the same time giving a broader and more sys- implementation, using instruments.
tematic perspective. Distinguishing between the political-administrative
Policy design, in terms of selecting and applying process of policy and instrument development, and the
instruments, is not a mechanical procedure with results policy instruments being used in the policy implemen-
independent of the wider social context and indepen- tation process is sometimes straightforward. The US,
dent of the qualities of the actors involved. There have for instance, has enacted laws on tradable emission
been simple descriptions of the policy process which permits for SO2 emissions, after lengthy research on
assume that the legislator enacts what is best, after how this instrument could function and lengthy politi-
which the regulations are implemented by law-abiding cal discussions on its advantages and disadvantages in
officers, leading to the intended effects, provided of terms of efficiency, effectiveness and ease of imple-
course that the technical preparations for legislation mentation. Implementation is a largely administrative
had been done properly. Political scientists have long process, upheld by checks and balances in which self-
since shown (see e.g. Easton 1965) how at a systems regulation plays a key role. Since nobody wants their
level, policy-making is related to political support, competitors to have a free ride, all trading parties sup-
limiting options for politicians and making the expected port officials checking the outcomes of emission trade,
outcome of regulations only one aspect in the process. especially if focussed on their competitors. The instru-
Sociologists of law have shown that similar laws work ment is clearly differentiated from its broader social
out differently depending on the administrative and and political context. With other instruments, however,
13 Environmental Policy Instruments 277

the distinction is not so clear. Covenants between gov- quality of nature. To these may be added amenity
ernments and groups of firms may be viewed from aspects, distributional aspects, the nature of our relation
different angles. In some instances, they create the with the biotic and abiotic environment and other nor-
discussion platform leading to concrete actions, as in mative aspects. The evaluation criteria specified in a
the Dutch packaging covenant. In this sense, it is not previous section may serve as a guide.
an instrument but a procedure that may lead to instru- Since empirical analysis is often very scanty in
mental use, if needed. Or covenants may already spec- environmental matters, it may be necessary to fall back
ify concrete actions for specific parties, as is also the on not fully proven assumptions. A striking conclu-
case in the Dutch packaging covenant. The ‘shadow of sion from research on the effectiveness of voluntary
the law’ is very explicitly present in this covenant, approaches is that little is known about the functioning
which states that it replaces direct regulation and, if of such approaches (cf. Harrison 1999). Their effec-
not successful, will be followed by more direct regula- tiveness has not been thoroughly studied, and where it
tion. The threat, of course, only works if regulatees – is assumed, this seems largely a matter of belief, simi-
in this case the private partners to the covenant contract lar to the old belief that binding instruments would
– expect government to be able to come up with this automatically lead to the effects specified. This may
legislation if deemed necessary. have been the case to some extent in the US, but in
Voluntary approaches (including voluntary agree- most European countries, there is a well-known gap
ments) cover procedural variants of regulatory instru- between legislation and implementation. In addition,
ments, as well as most social instruments. They are to legislation may enact what would have happened any-
be distinguished from the social procedures followed way. This safeguards effectiveness in terms of being in
in the preparation of other types of regulatory instru- line with legislation, while effectiveness of policy in
ments, like permits or emission taxes, which them- terms of a behavioural adjustment for environmental
selves are not voluntary. The difference is not always improvement may be more or less lacking. Similarly,
clear though, as in the example of ‘permit preparation’, instrument debates often state that tradable emission
which was formerly a major instrument for environ- permits lead to definite environmental effects, unlike
mental policy in England and the Netherlands. emission taxes. Let us take CO2 emissions from elec-
tricity production as an example. If, due to unusually
hot and cold weather periods, electricity consumption
[Link] Instrument Analysis for Evaluation and related CO2 emissions have eaten up the available
permits, it can hardly be imagined that electricity pro-
Evaluating specific policy instruments requires a com- duction will therefore be shut down for the last half of
bination of normative and empirical analysis. The nor- December. Seemingly rigid instruments are not applied
mative analysis guides the empirical analysis, as only to the full, while soft instruments like taxes may have
results that are relevant in normative terms are relevant very predictable results at an aggregate level.
to the evaluation. As usual, however, things are not as This state of affairs in empirical analysis for evalu-
simple in practice as they are in principle. In empirical ation may be discomforting but should not lead to iner-
terms, two types of mechanism are involved in the tia. In real life, as opposed to science, a best guess is
effect route towards environmental policy aims or, better than none at all, and defective but comprehen-
using a broader definition, sustainability aims. The first sive evaluation schemes are to be preferred to doing
group includes mechanisms in society, the second those nothing, or to focusing policy on some partial effects
in the environment. For both types of mechanisms one because other things have not been fully proven. If a
may distinguish between primary effect mechanisms, balance is to be struck, one at least needs to know what
essentially reducing causalities to one single chain, and is not fully proven, and to see where evaluation prob-
secondary mechanisms, involving feedback loops, lems reside. The scheme of Fig. 13.2 at least indicates
modelled in simple or more advanced ways. In norma- a number of relevant types of information. An analysis
tive terms, there is no well-structured set of values that and evaluation which may be faulty in some respects is
can be linked to environmental policy. There is a pos- better than none at all, and policy instruments should
sible classification, however, into major value fields preferably be set up in a way which best reflects available
related to human health, economic prosperity and the knowledge, limited though it may be.
278 G. Huppes and U.E. Simonis

[Link] Strategic Considerations Above emission types are made, they define a number of
the Case Level equations (usually of the type ax = y). An example is
the global warming effect as empirically modelled for
We now return to strategic aspects of instrument choice various substances, for instance with 20 units of car-
and evaluation going beyond the design and evaluation bon dioxide (ax) equalling 1 unit of Global Warming
of specific problem-and-instrument combinations. One Potential for methane (y). By eliminating the specific
question is whether a specific problem can best be substances and their equations, the number of goals is
solved by one or a few instruments, as opposed to a reduced to one: global warming (more precisely: time-
broad mixture of instruments. Another question is integrated climate forcing). This integration can be
whether instrument choice is to focus on a limited set of pursued further, on a normative basis, for instance by
instruments for environmental policy, as opposed to a integrating the overall toxicity scores of all toxic emis-
set that is as broad as possible in order to fit all purposes. sions with this total global warming effect of all
Piecemeal improvement at the micro level of individual climate-changing substances, reducing them to one
instruments may well be sub-optimal at the macro level evaluation-based denominator. The level of aggrega-
of societal environmental policy as a whole. Instrument tion defines the problem, in that this discussion on
choice in the sense of adding new instruments may not numbers of instruments and goals may hold for a given
simply be a matter of finding the best options for the problem level.
environmental problem case at hand. Both of these gen- Before accepting these quite stringent conclusions
eral strategic questions relate back to instrument choice on instrument numbers equalling the number of goals,
at the operational level as discussed above. however, it should be examined if this line of reason-
The first question actually asks whether there is a ing is really applicable. There are several reasons why
minimum, optimum and maximum number of instru- this does not seem to be the case. Some are based on
ments; in other words, it asks how sparsely instruments the nature of goals in environmental policy, others
should be used and which level of variety is necessary relate to the somewhat loose nature of instrument and
and wise. In the context of macro-economic policy, goal specification in relation to the underlying dirty
there is a clear preference for using Occam’s razor: the and fuzzy models of the world. It may also be ques-
minimum number of instruments required is also the tioned whether the goals are really discrete values or
maximum number to be used. There is a basic logic in are of the nature of variables to be maximised, for
mathematical models that says that the number of inde- which overshooting the mark is no problem at all,
pendent variables, that is, the instruments, has to be merely yielding a slightly higher environmental qual-
equal to the number of dependent variables, the goals ity than originally expected or intended. If this is true,
or aims (see Tinbergen 1967). This holds for a wide the problem should be seen as a maximisation (or in
range of model types, but most clearly for linear mod- this case minimisation) problem with boundary condi-
els. The reasoning is different for the two types of pos- tions, rather than as a solution to arrive at a discrete
sible deviations: fewer or more. A smaller number of value. This implies that the requirement of equality
instruments will not allow the goals to be achieved, and between numbers of instruments and numbers of goals
non-effectiveness is obviously a mortal sin in instru- vanishes, as is generally accepted in administrative
ment design. If there are more instruments than goals, sciences (see e.g. Gunningham and Sinclair 1999).
as is often advised, the system has an infinite number of Different instruments may each take care of particular
solutions. The value of the instrument variables can improvements, possibly supporting each other in their
only be derived by arbitrarily reducing the number of functioning, in that for instance an informational
instrument variables to the point where the remaining instrument may be conducive to the active develop-
number once more equals that of the goals. This could ment of permits. Furthermore, the relation between
hardly be called a well-founded procedure. It would instruments and goals may not always be so clear as to
probably be more rational and adequate if, based on imply a one-to-one link. In most real life situations,
considerations external to the model, a selection of there is no clear model with instruments put in at one
instruments could be made beforehand. end and goal achievement pouring out at the other. The
The discussion on policy integration is closely relation between instruments and goals is not so
related to this subject. If trade-offs between different straightforward. Ultimately, each instrument is at least
13 Environmental Policy Instruments 279

to some extent linked to all environmental problems, Sweeney (Eds.), Handbook of natural resource and energy
albeit in widely different degrees and with very large economics. Amsterdam: North Holland.
Bonus, H. (Ed.). (1998). Umweltzertifikate. Der steinige weg zur
uncertainties. Also, instruments may be mutually marktwirtschaft (sonderheft 9/1998 der zeitschrift für ange-
supportive. The underlying models for these instru- wandte umweltforschung). Berlin: Analytica.
ments then differ somewhat in scope. A more practical Bressers, J. T. A. (1988). A comparison of the effectiveness of
approach would be to see them as parallel applications incentives and directives: The case of the Dutch water qual-
ity policy. Policy Studies Review, 7(3), 500–518.
of several instruments. Bressers, J. T. A., & Klok, P. J. (1988). Fundamentals for a the-
Finally, some types of instruments belong together ory of policy instruments. International Journal of Social
and may be mutually supportive, while others are con- Economics, 15(3/4), 22–41.
flicting. In environmental policy, there has always been a Brundtland Report: World Commission on Environment and
Development. (1987). Our common future. Oxford: Oxford
tension between the control-oriented permitting sys- University Press.
tem and technology-binding and behaviour-prescriptive Castells, M. (1996). The rise of the network society. Oxford:
approaches on the one hand, and liability and taxing Blackwell.
approaches on the other. A great deal of detailed legal Castells Cabré, N. (1999). International environmental agree-
ments. Institutional innovation in European transboundary
sophistication has developed to reconcile these conflict- air pollution policies. Amsterdam: Free University Press
ing approaches in regulatory practice. Introducing the (doctoral dissertation).
tradable emission permit, which is already operational in Coase, R. H. (1960). The problem of social cost. Journal of Law
the US on sulphur dioxide emissions and is now being and Economics, 3(1), 1–44.
Cole, D. H., & Clark, J. (1998). Poland’s environmental trans-
considered for climate changing emissions in Europe, formation: An introduction. In J. Clark & D. H. Cole (Eds.),
leads to a substantial shift in this balance, for one thing Environmental protection in transition. Economic, legal and
because carbon dioxide emissions are intimately linked socio-political perspectives on Poland. Aldershot: Ashgate.
to all economic activities. The second approach, follow- Easton, D. (1965). A systems analysis of political life. Englewood
Cliffs, NJ: Prentice-Hall.
ing the pure version of the polluter-pays principle that Endres, A. (1997). Incentive-based instruments in environmen-
was worked out by the OECD three decades ago, is com- tal policy: Conceptual aspects and recent developments.
promised by the advent of tradable emission permits. It Konjunkturpolitik, 43(4), 298–343.
will become quite impossible to shift to such a taxing EPA. (1999). Description of 9 proposed sites and 10 final sites
added to the national priorities list in October 1999.
system once the tradable permit system has been estab- Intermittent Bulletin, 2(5) [online].
lished, whereas such a shift is still possible whilst tech- Fischer-Kowalski, M. et al. (1997). Gesellschaftlicher Stoffwechsel
nologies are being regulated with permits. Such basic und Kolonisierung von Natur. Amsterdam: Verlag Fakultas.
institutional choices, which need to be made explicitly, Galanter, H. (1981). Justice in many rooms: Courts, private ordering
and indigenous law. Journal of Legal Pluralism, 19, 1–47.
would have to be based on a rethinking of environmental Giddens, A. (1994). Beyond left and right (p. 15, 191).
regulation in particular, of public regulation in general, Cambridge: Polity Press.
and of the even more general principles of social organi- Giddens, A. (1998). The third way: The renewal of social democ-
sation. The outcome of this broader societal analysis and racy. Cambridge: Polity Press.
Guinée, J. B., Gorrée, M., Heijungs, R., Huppes, G., Kleijn, R., de
decision-making process could help guide all choices on Koning, A., et al. (2002). Handbook on life cycle assessment.
instruments for environmental policy. Operational guide to the ISO standards. Dordrecht: Kluwer
(now Springer).
Gunningham, N., & Sinclair, D. (1999). Regulatory pluralism:
Designing policy mixes for environmental protection. Law
References and Policy, 21(1), 49–76.
Harrison, K. (1999). Talking with the donkey: Cooperative
approaches to environmental protection. Journal of Industrial
Bardach, E., & Kagan, R. A. (1982). Going by the book; the Ecology, 2(3), 51–72.
problem of regulatory unreasonableness. Philadelphia: Hawkins, K. (1984). Environment and enforcement. Regulation and
Temple University Press. the social definition of pollution. Oxford: Clarendon Press.
Baumol, H., & Oates, W. E. (1988). The theory of environmental Hemmelskamp, J. (1997). Environmental policy instruments
policy. Cambridge: Cambridge University Press. and their effects on innovation. European Planning Studies,
Beck, U., & Beck-Gernsheim, E. (2002). Individualization: 5(2), 177–194.
Institutionalised individualism and its social and political Hoffman, A. J. (2006). Getting ahead of the curve: Corporate
consequences. London: Sage. strategies that address climate change. PEW centre on cli-
Bohm, P., & Russell, C. S. (1985). Comparative analysis of mate change. See chapter on shell. Ann Arbor, MI: University
alternative policy instruments. In A. V. Kneese & J. L. of Michigan.
280 G. Huppes and U.E. Simonis

Huppes, G. (1988). New instruments for environmental policy: (Eds.), Climate change policy: A survey. Washington, DC/
A perspective. International Journal of Social Economics, London: Island Press.
15(3/4), 42–50. Schumpeter, J. A. (1976). Capitalism, socialism and democracy.
Huppes, G. (1993). Macro-environmental policy, principles and London: George Allen & Unwin.
design. Amsterdam: Elsevier. Sen, A. K. (1977). Rational fools. A critique of the behavioral
Huppes, G., & Kagan, B. (1989). Market-oriented regulation of foundations of economics. Philosophy and Public Affairs, 6,
environmental problems in the Netherlands. Law and Policy, 317–344.
11(2), 215–239. Sen, A. K. (1982). Choice, welfare and measurement. Cambridge,
Huppes, G., van der Voet, E., van der Naald, W. G. H., MA/London: Harvard University Press.
Vonkeman, G., & Maxson, P. (1992). New market-oriented Sen, A. K. (2000). Consequential evaluation and practical rea-
instruments for environmental policies. London: Graham & son. Journal of Philosophy, 97(9), 477–503.
Trotman. Simonis, E. U. (1998). Internationally tradable emission certifi-
Huppes, G., Sas, H., de Haan, E., & Kuyper, J. (1997). Efficiënte cates: Efficiency and equity in linking environmental protec-
milieuinvesteringen (efficient environmental investments). tion with economic development. In H. -J. Schellnhuber & V.
Milieu, 12(3), 126–133. Wenzel (Eds.), Earth system analysis: Integrating science
Imhoff, M. L., Bounoua, L., Ricketts, T., Loucks, C., Harriss, for sustainability: Completed results of a symposium orga-
R., & Lawrence, W. T. (2004). Global patterns in human nized by the Potsdam Institute. Berlin: Springer.
consumption of net primary production. Nature, 429, Superfund Programme. (1986). This programme has been adopted
870–873. in two steps. CERCLA (Comprehensive Environmental
Ingold, T. (1986). The appropriation of nature: Essays on human Response, Compensation and Liability Act) in 1980 and
ecology and social relations. Manchester: Manchester SARA (Superfund Amendments and Reauthorisation Act) in
University Press. 1986.
Jänicke, M., & Jorgens, H. (1998). National environmental policy van Tatenhove, J., & Leroy, P. (2000). The institutionalisation of
planning in OECD countries: Preliminary lessons from cross- environmental politics. In J. van Tatenhove, B. Arts, & P.
national comparisons. Environmental Politics, 7(2), 27–54. Leroy (Eds.), Political modernisation and the environment:
Luhmann, N. (1989). Ecological communication. Cambridge: The renewal of environmental policy arrangements.
Polity Press. Dordrecht: Kluwer.
Mayntz, R. (1978). Volzugsprobleme der umweltpolitik (applica- von Benda-Beckmann, K., & Hoekema, A. J. (1987). Hoe hori-
tion problems in environmental policy). Stuttgart: Kohlhammer. zontaal bestuurt de overheid (how horizontal is govern-
Mintzberg, H. (1994). The rise and fall of strategic planning. ment?). Recht der Werkelijkheid, 8(2), 3–12.
New York: Free Press. von Neumann, J., & Morgenstern, O. (1953). Theory of games
OECD (Organisation for Economic Co-operation and and economic behavior. Princeton, NJ: Princeton University
Development). (1995). Climate change. Economic instru- Press.
ments and income distribution. Paris: OECD. Tinbergen, J. (1967). Economic policy: Principles and design:
Opschoor, J. B., & Vos, H. B. (1989). The application of economic See Section 3.3. Amsterdam: North-Holland.
instruments for environmental protection. Paris: OECD. Tisdell, C. (1998). Economic policy instruments and environ-
Rawls, J. (1972). A theory of justice. London: Oxford University mental sustainability: Another look at environmental-use
Press. permits. Sustainable Development, 6, 18–22.
Riesman, D. (1950). The lonely crowd: A study of the changing Tönnies, F. (1887). Gemeinschaft und gesellschaft: Abhundlung
American character. New Haven, CT: Yale University des communismus und des socialismus als empirischer cul-
Press. turformen. Leipzig: Fues’s Verlag.
Scharpf, F. W. (1991). Die handlungsfähigkeit des staates am ende Vogel, D. (1986). National styles of regulation: Environmental
des zwanzigsten jahrhunderts. Politische Vierteljahresschrift, policies in Great Britain and the United States. London:
32(4), 621–634. Cornell University Press.
Schneider, S. H., & Kuntz-Duriseti, K. (2002). Uncertainty and Weber, M. (1947). The theory of economic and social organisation.
climate change policy. In S. H. Scheider & A. Rosencranz New York: Oxford University Press.
Chapter 14
Environmental Institutions and Learning: Perspectives
from the Policy Sciences

Matthijs Hisschemöller, Jan Eberg, Anita Engels, and Konrad von Moltke†

Contents 14.1 Political Institutions, Knowledge

14.1 Political Institutions, Knowledge
and Power
and Power ............................................................... 281
14.2 Environmental Policy Analysis This chapter on environmental policy institutions and
and Learning .......................................................... 284 learning opens with a brief introduction on the problems
14.2.1 Institutionalism and dilemmas political institutions face in dealing with
and Constructivism .................................................. 284
14.2.2 Conceptualizing
knowledge and power. The reason is twofold: on the one
the Policy Process .................................................... 284 hand, such an introduction provides a framework which,
14.2.3 The Advocacy Coalition in our view, is critical for an adequate understanding of
Framework ............................................................... 285 the challenges for environmental institutions and the
14.2.4 Policy-Oriented Learning......................................... 286
14.2.5 The ACF and Environmental
policy science’s concepts, theories and methods to
Policy ....................................................................... 287 investigate and advise with respect to these institutions.
On the other hand, the study of environmental institu-
14.3 Institutionalisation of Policy-Science
Interfaces and Environmental Policy ................... 288 tions and the way they may adapt to the needs of society
14.3.1 Different National Traditions ................................... 289 may provide valuable lessons for the current discussions
14.3.2 International Assessment ......................................... 291 on political institutions in general.
14.4 Biases with Respect to Stakeholder In this chapter, the concept institution is defined in a
Participation in Environmental somewhat different way than may happen in everyday
Policy Institutions................................................... 292 speech, where the concept is often used as identical to
14.4.1 Policy as Rule .......................................................... 294
14.4.2 Policy as Negotiation ............................................... 295
organisations. In the social sciences, institutions refers
14.4.3 Policy as Accommodation ....................................... 295 primarily to formal and informal ‘rules of the game’
14.4.4 Policy as Learning.................................................... 295 that shape individual and group behaviour. Defined in a
14.5 Methods to Enhance broad sense, this may include organisations as well, but
Stakeholder Participation ..................................... 296 institutions may exist with and without organisations.
14.6 Summary and Conclusions ................................... 300 Examples of political institutions are visible organisa-
tions such as national states, the European Union or the
References ........................................................................... 302
United Nations. But equally important are less visible
though influential institutions such as markets, formal
and informal policy-science interfaces and laws.
One of the major problems for political philosophers
and practitioners, working on the design and evaluation
of political institutions, has been the question how such
institutions can do justice to the will of the people and
M. Hisschemöller ()
VU University Amsterdam, The Netherlands to scientific knowledge. For those, who did not find
e-mail: [Link]öller@[Link] themselves bound by democratic principles or even
†
Konrad von Moltke passed away in 2005 rejected democracy as mediocracy that would render

Jan J. Boersema and Lucas Reijnders (eds.), Principles of Environmental Sciences, 281
© Springer Science + Business Media B.V. 2009